PORPHYRIN COMPLEXATION: AN APPROACH IN PORPHYRIA …

PORPHYRIN COMPLEXATION:

AN APPROACH IN PORPHYRIA THERAPY

By

Bolanle C. AKINWUMI

A thesis submitted to the Faculty of Graduate Studies of

The University of Manitoba

in partial fulfilment of the requirements of the degree of

MASTER OF SCIENCE

Faculty of Pharmacy,

University of Manitoba,

Winnipeg, Manitoba, Canada

Copyright ® 2012 by Bolanle C. Akinwumi

Abstract

i

Abstract

Porphyria is a rare metabolic disease which occurs as a result of accumulation of

endogenous porphyrins due to specific enzyme deficiency in the biosynthetic pathway of

heme. Chloroquine is currently used in the treatment of cutaneous porphyria, although its

mechanism of action is not yet well understood. It is believed that chloroquine works in

porphyria by forming complexes with excess porphyrin molecules and thus enhancing

their elimination from the body. Previous reports of porphyrin-chloroquine complexes

have been done mostly in aqueous models. In this study, UV/Visible optical absorbance

difference spectroscopy was used to study the complexation of protoporphyrin IX with

chloroquine and a range of acceptor molecules in hydrophobic models. The results show

that chloroquine, mefloquine, amodiaquine, quinacrine, and pyronaridine formed

relatively stronger complexes compared to other molecules such as quinine, duroquinone

and caffeine. Therefore, relative to chloroquine, some of the molecules with comparable

or greater binding affinity to protoporphyrin IX might also be useful in the treatment of

porphyria.

Acknowledgements

ii

Acknowledgements

I want to acknowledge the support of my advisor Dr. Alan McIntosh for his

encouragement during the course of my research program. His knowledge and

enthusiasm for science is always an inspiration to me.

I also appreciate the unrelenting efforts of my advisory committee members: Dr. Cindy

Ellison, Dr. Frank Burczynski, and Dr. Silvia Alessi-Severini. Thank you all for your

kind words of guidance. I feel highly honored to have had you by my side through the

journey.

This work would not be a success without the generous financial support received

through the following scholarships: Pfizer Canada Centennial Research Award, Faculty

of Graduate Studies Special award, and the Leslie F. Buggey Scholarship award in

Pharmacy.

Special thanks also go to my friends and family both in Canada and back home in Nigeria

for their support during the course of my program, especially my fiancé and my siblings.

Dedication

iii

Dedication

This project is dedicated to my parents: Dr. and Mrs. Isaac Akinwumi for their

unwavering support from my childhood till date.

Table of contents

iv

Table of Contents

Abstract ………………………………………………………………………………. i

Acknowledgements …………………………………………………………………… ii

Dedication …………..…………………………………………………………………. iii

Table of Contents……………………………………………………………………….. iv

List of Tables ………………………………………………………………………….. vii

List of Figures ………………………………………………………………...……….. viii

List of Equations………….………………………………………………..…………. xi

List of Abbreviations ………………………………………………………………….. xii

CHAPTER I: INTRODUCTION

1.0 Background ..………………………………………………………………….. 1

1.1 Porphyria disease ...………………………………………………….. 1

1.2 Heme biosynthesis and regulation ……………………………………. 2

1.3 Classification of porphyrias ....…….……………………………………. 5

1.4 Pathogenesis of porphyrias ……………………………………....…. 7

1.5 Diagnosis of porphyrias ……………………………………. 9

1.6 Current Clinical trials ……………...…………………………… 10

1.7 Treatment options and mechanisms ….……………………..………… 11

1.7.1 Acute porphyrias …….…………………………..………… 11

1.7.2 Cutaneous porphyrias …………………………...……… 12

2.0 Literature review …………………………………………………..………. 13

2.1 In vivo Localization and excretion of porphyrins ……………………… 13

2.2 Chloroquine use in Porphyria therapy .………………….………. 14

2.2.1 Chloroquine mechanism of action in Porphyria .………..… 15

2.2.2 Chloroquine -antimalarial mechanism of action ...…………. 17

2.3 Quantitative characterization of chloroquine-porphyrin complexes ….…17

Table of contents

v

2.3.1 Aqueous models .…………………………………..……… 17

2.3.2 Need for Hydrophobic models …………………….…….. 19

2.4 Protoporphyrin IX molecule …………………………………...……… 20

2.5 Molecular interactions: binding and complexes ……………..……. 21

3.0 Other compounds to be tested …………………………………….…….. 24

3.1 Chloroquine related compounds ……….………………………. 24

3.2 Simple ring systems …………………………….….... 25

3.3 Other ring systems …………………………………. 25

4.0 Experimental design and approach to be employed ..……….………….……. 27

4.1 Hydrophobic models …………………………………. 30

4.2 Graphical and mathematical models ...………………………………. 32

CHAPTER II: RATIONALE AND HYPOTHESES

1.0 Rationale for the research …………….……………………..………… 36

1.1 Hypotheses of the research …………………………………………. 38

1.2 Objectives of the research …………………………………….……. 38

CHAPTER III: MATERIALS AND METHODS

1.0 Materials ………………………………………….. 39

2.0 Acetone: Dichloromethane 50:50 solvent system ………………..…………… 40

2.1 Preparation of stock solutions ……………….…...……………………. 40

2.2 Extraction of free base ………...…………………………….….. 41

2.3 Complexation mixtures in homogenous solutions containing PPIX and

each acceptor molecule……………………………………………..….. 43

2.4 UV-Visible difference spectroscopic analysis …………………….…… 43

3.0 Triton X-100 aqueous micellar systems ………………………………..……… 46

4.0 Bovine Serum Albumin system …………………………………..…..... 48

Table of contents

vi

5.0 Graphical and Mathematical models and statistical analysis ………….……….. 50

5.1 Benesi-Hildebrand Equation and Plots ……………………….….. 50

5.2 Binding curves ………………………………...… 50

5.3 Hill and Scatchard plots …………………………………….…….. 50

CHAPTER IV: RESULTS AND DISCUSSION

1.0 Acetone: Dichloromethane 50:50 homogenous solvent system ……..………… 51

1.1 Weak association category ….………………….………….……. 52

1.1.1 Duroquinone …………………………….…….. 52

1.1.2 Quinine ………...………………….……... 53

1.1.3 Caffeine ……………………….………….. 54

1.2 Strong association category ……………………….………….. 58

1.2.1 Difference spectra …………………….…….………. 58

1.2.2 Binding curves ……………………….………….. 65

1.2.3 Scatchard plots ………………………….……….. 70

1.2.4 Hill plots ………………………….……….. 74

1.3 Molecular interactions and important structural features …………..…. 78

1.4 Unique PPIX complexes with other acceptor molecules: amitriptyline and

chlorpromazine ………………………..…………….….. 83

2.0 Triton X-100 model ……………………………………….….. 87

3.0 Bovine serum albumin ……………………………………….….. 89

CHAPTER V: CONCLUSIONS AND RECOMMENDATIONS

1.0 Conclusions ……………………………….………….. 94

2.0 Recommendations ………………………………….……….. 96

REFERENCES ………………………………………………………….……….. 97

List of tables

vii

List of tables

Table 1: Classification of the porphyrias ……………………………………………… 5

Table 2: Table of chemicals ……………………………………......…... 39

Table 3: Dissociation constants for molecules in the weak association category …..…. 55

Table 4: Dissociation constants for molecules in the strong association category …...... 77

Table 5: pKa values of the acceptor molecules (most basic nitrogens) ………….……78

List of Figures

viii

List of Figures

Figure 1: Heme biosynthesis and the porphyrias ……………………...……………. 3

Figure 2: Last step in the biosynthesis of heme ……………………...……………. 4

Figure 3: Pathogenesis of cutaneous effects in porphyria …………………………...….. 8

Figure 4: Theoretical superimposition of protoporphyrin IX and chloroquine structures in

a 1:1 complex ……………………………………..… 22,79

Figure 5: Structures of electron acceptors under study …………………………...……. 26

Figure 6: Spectrum of protoporphyrin IX (36μM) showing the four Q-bands ……...…. 27

Figure 7: Spectral changes as expected in the protoporphyrin IX spectrum ……...... 28,51

Figure 8: Illustration of Beer Lambert Law ………...………………………………… 29

Figure 9: Triton X-100® structures ………………………….…………..... 30,87

Figure 10: Plots showing background absorption of acceptor molecules ……………… 45

Figure 11: Difference spectra for duroquinone-PPIX complexes, and duroquinone

structure …………………………………………………... 52

Figure 12: Difference spectra for quinine-PPIX complexes and quinine structure……. 53

Figure 13: Difference spectra for caffeine-PPIX complexes and caffeine structure……54

Figure 14: Benesi-Hildebrand plots for duroquinone, quinine and caffeine …………... 55

Figure 15: Difference spectra for chloroquine-PPIX complexes and chloroquine

structure…………………………..……………………………………………...…….... 60

Figure 16: Difference spectra for amodiaquine-PPIX complexes and amodiaquine

structure………………………………………………………………………………..... 61

Figure 17: Difference spectra for mefloquine-PPIX complexes and mefloquine

structure………………………………………………………………………………..... 62

Figure 18: Difference spectra for quinacrine-PPIX complexes and quinacrine

structure………………………………………………………………………………..... 63

Figure 19: Difference spectra for pyronaridine-PPIX complexes and pyronaridine

structure…………………………………………………………………………………. 64

Figure 20: Binding curves for chloroquine at four different wavelengths ……………... 66

List of Figures

ix

Figure 21: Binding curves for amodiaquine at four different wavelengths (n=3) ………67

Figure 22: Binding curves for amodiaquine at four different wavelengths (n=6) ………67

Figure 23: Binding curves for mefloquine at four different wavelengths ……………....68

Figure 24: Binding curves for quinacrine at four different wavelengths ………………..69

Figure 25: Binding curves for pyronaridine at four different wavelengths ……………..69

Figure 26: Scatchard plots for chloroquine at 516nm in triplicate measurements …...…72

Figure 27: Scatchard plots for amodiaquine at 521 nm in triplicate measurements……..72

Figure 28: Scatchard plots for quinacrine at 535nm in triplicate measurements………...73

Figure 29: Scatchard plots for mefloquine at 516 nm in triplicate measurements………73

Figure 30: Hill plot for chloroquine (516nm) …………………………………………....74

Figure 31: Hill plots for amodiaquine (521nm) (n=3 determination) …………………...75

Figure 32: Hill plot for amodiaquine (521nm) (n=6 determination) ……………………75

Figure 33: Hill plots for mefloquine (516nm) and quinacrine (535nm) ………….……..76

Figure 34: Theoretical superimposition of protoporphyrin IX and quinacrine structures in

a 1:1 complex …………………………………………………………………... 80

Figure 35: Theoretical superimposition of protoporphyrin IX and amodiaquine structures

in a 1:1 complex …………………………………………………………………... 80

Figure 36: Theoretical superimposition of protoporphyrin IX and mefloquine structures in

a 1:1 complex ………………………………………………………………….. 81

Figure 37: Theoretical superimposition of protoporphyrin IX and pyronaridine structures

in a 1:1 complex …………………………………………………………………... 82

Figure 38: Superimposition of PPIX and chlorpromazine in a possible electrochromic

interaction …………………………………………………………………... 84

Figure 39: Difference spectra for amitriptyline-PPIX complexes and amitriptyline

structure …………………………………………………………………... 85

Figure 40: Difference spectra for chlorpromazine-PPIX complexes and chlorpromazine

structure …………………………………………………………………... 86

Figure 41: Difference spectra for amodiaquine-PPIX in triton X-100® …….………… 88

List of Figures

x

Figure 42: Difference spectra for chloroquine-PPIX in triton X-100® ………………....88

Figure 43: PPIX spectrum (36µM) in the presence of BSA …………………………….90

Figure 44: PPIX Spectrum (36µM) in acetone/dichloromethane solvent system …........90

Figure 45: Spectrum of chloroquine in the presence of BSA at pH 7.40 and pH 9.00…..92

List of equations

xi

List of Equations:

Equation 1: Modification of Benesi-Hildebrand equation …………………………..…. 31

Equation 2: Hill equation ………………………………………………..…. 32

Equation 3: Scatchard equation ……………………………………………..……. 33

List of Abbreviations

xii


AIP Acute intermittent porphyria

ALA Aminolevulinic acid

AMQ Amodiaquine

B-H Benesi-Hildebrand

BSA Bovine serum albumin

CAF Caffeine

CPZ Chlorpromazine

CQ Chloroquine

DDC 3,5-Dicarbethoxy-1,4-dihydro-2,4,6-trimethylpyridine

DQ Duroquinone

EMA European Medicines Agency

EPP Erythropoietic protoporphyria

FDA Food and Drug Administration

HCB Hexachlorobenzene

Ka Association constant

Kd Dissociation constant

MFQ Mefloquine

NMR Nuclear magnetic resonance

PBG Porphobilinogen

PCT Porphyria cutanea tarda

PEG Polyethylene glycol

PPIX Protoporphyrin IX

PYR Pyronaridine

QCR Quinacrine


xiii

QN Quinine

ROS Reactive oxygen species

Tris Tris(hydroxymethyl)aminomethane

UV Ultraviolet

ΔA Change in absorbance

Δ-ALAS Delta-aminolevulinic acid synthase

Π-П Pi to pi

Introduction

1

CHAPTER I: INTRODUCTION

1.0 Background:

The word porphyria derives from the Greek word ‘porphyra’, meaning purple pigment.

When specific enzymes are deficient in the biosynthesis of heme, there is accumulation

of toxic intermediates called porphyrins in the body, mainly in the blood and liver cells.

Excess porphyrins are excreted in the urine and feces (stained purple in colour) of

patients suffering from the disease. Administration of drug molecules that can form

complexes with the excess porphyrins may enhance their removal from the body either

through the urine or via the enterohepatic route thus appearing in the feces.

Exploitation of porphyrin complexation as a mechanism in porphyria therapy requires

background knowledge of the chemical interactions between the porphyrins and potential

drug molecules. This research is therefore focussed on studying the complexation

reactions between one of the porphyrin molecules and some selected compounds that can

be employed in the treatment of porphyria. In order to fully appreciate the significance

and application of this study, it is necessary first to understand the nature of the porphyria

disease state, its pathogenesis and treatment options and also to review the few works that

have been carried out previously in this research area.

1.1 Porphyria disease:

The porphyrias are a complex group of metabolic genetic diseases with 8 different

clinical manifestations that correspond to different enzyme deficiencies in the

biosynthesis of heme 1. Heme is a component of hemoproteins that serve various

functions in the body e.g. hemoglobin, myoglobin, cytochrome P450 and cytochrome C.

Hemoglobin and myoglobin are responsible for oxygen transport and storage

Introduction

2

respectively. The cytochrome P450 enzymes form an important group of monooxygenase

enzymes responsible for the oxidative phase I metabolism of most drugs in the liver and

in other organs. Lastly, cytochrome C is involved in oxidative phosphorylation in the

mitochondria of most body cells. Heme is therefore an essential molecule in the body.

1.2 Heme biosynthesis and regulation:

Heme is synthesized in a cascade of multi-step reactions (Figure 1) mainly by the

erythroid cells in the bone marrow and also by the liver cells 1. As shown in Figure 1,

each porphyria subtype corresponds to a specific enzyme deficiency occurring at a

specific stage of the biosynthetic pathway. The first step, which is the rate limiting step,

is the formation of δ-aminolevulinic acid (ALA) from the condensation of glycine and

succinyl coenzyme A catalyzed by the enzyme δ-aminolevulinic acid synthase (δ-ALAS).

The first biosynthetic step along with the last three steps occurs in the mitochondria

where their respective enzymes are located. However, the other steps take place in the

cytosol. Protoporphyrin IX (PPIX) is the last intermediate in the biosynthetic pathway. In

the last step of heme synthesis, the enzyme ferrochelatase inserts an iron (II) atom into

the centre of the tetrapyrrole ring of the protoporphyrin IX molecule to form heme

(Figure 2).

Introduction

3

Figure 1: Heme biosynthesis and the porphyrias

Heme

Protoporphyrin IX

Ferrochelatase Erythropoietic porphyria

Protoporphyrinogen IX

Protoporphyrinogen oxidase Variegate porphyria

Coproporphyrinogen III

Coproporphyrinogen oxidase Hereditary coproporphyria

Uroporphyrinogen III

Uroporphyrinogen decarboxylase Porphyria cutanea tarda

Hydroxymethylbilane

Uroporphyrinogen III synthase Congenital erythropoietic porphyria

Porphobilinogen

Porphobilinogen deaminase Acute intermittent porphyria

Deta aminolevulinic acid

Aminolevulinate dehydratase ALA dehydratase deficiency porphyria

Glycine + Succinyl-CoA

Delta aminolevulinic acid synthase

Introduction

4

Figure 2: Last step in the biosynthesis of heme

In the liver, heme biosynthesis is regulated by a negative feedback mechanism by excess

heme, which represses the transcription of the gene for the δ-ALAS enzyme, thus,

inhibiting the rate limiting step of the biosynthetic pathway. However, in the bone

marrow, the rate of synthesis is limited only by iron availability –uptake of iron. That is,

heme biosynthesis proceeds as long as iron is available in the erythroid cells usually

during their differentiation stage in the bone marrow 1. Heme is degraded by the action of

the heme oxygenase enzyme to form biliverdin and then further converted to bilirubin by

biliverdin reductase. Bilirubin is excreted via the enterohepatic route while iron is

recycled for use in heme synthesis 2.

Introduction

5

1.3 Classification of Porphyrias:

Porphyria subtypes can be broadly classified based on clinical manifestations such as the

acute porphyrias and cutaneous porphyrias (Table 1)

Table 1: Classification of the porphyrias

Hepatic Porphyrias

Acute ALA-Dehydratase – Deficiency Porphyria

Hereditary Coproporphyria

Acute Intermittent Porphyria (AIP)

Cutaneous Porphyria Cutanea Tarda (PCT)

Variegate Porphyria

Erythropoietic Porphyrias

X-Linked Sideroblastic anemia

Erythropoietic Protoporphyria (EPP)

Congenital Erythropoietic Porphyria

Acute porphyrias- Signs and Symptoms: neurologic effects could be autonomic such as

severe abdominal pain; peripheral such as muscle weakness; or central such as loss of

consciousness, anxiety, hallucination, and depression 1. The central nervous effects in

porphyrias have often been confused with other neurological disorders.

Introduction

6

Cutaneous porphyrias- Signs and symptoms: burning, blistering, and edema of the skin

on exposure to sunlight. These skin reactions are usually associated with debilitating

pain.

Porphyrias can also be classified based on the site of expression of deficient enzymes

such as the hepatic and erythropoietic porphyrias (Table 1). The hepatic porphyrias are

mostly associated with acute neurologic manifestations while the erythropoietic

porphyrias often present with cutaneous reactions on exposure to light.

Despite this broad classification into hepatic and erythropoietic porphyrias, some hepatic

porphyrias are also characterized by cutaneous manifestations while certain cutaneous

porphyrias may have some hepatic involvement. Overlapping of clinical presentations

and of the signs and symptoms among different porphyria classes makes a specific

diagnosis difficult.

Introduction

7

1.4 Pathogenesis of Porphyrias:

Accumulation of heme precursors (mostly porphyrins) that occurs as a result of specific

enzyme deficiencies is the main cause of most clinical manifestations of the disease.

Cutaneous Porphyria: In cutaneous porphyrias, excess porphyrins move out of the blood

cells and liver and accumulate in the skin where they cause two forms of photosensitivity

reactions: delayed blistering/scarring in porphyria cutanea tarda (PCT) and immediate

burning pain as in erythropoietic protoporphyria (EPP). On exposure to sunlight, the

porphyrin molecule absorbs light energy and becomes converted into the singlet excited

state (Figure 3). The singlet state porphyrin has a very short life time because it rapidly

changes to the lower energy but highly reactive triplet state porphyrin via a process called

intersystem crossing. While in the triplet state, the porphyrin molecule can react with

ground state molecular oxygen in vivo leading to the generation of reactive oxygen

species (ROS) such as singlet oxygen and hydroxyl radicals. The toxic ROS then react

with biomolecules in the skin to cause damage through destructive processes such as lipid

peroxidation and membrane and protein oxidation 3.

Acute/Hepatic Porphyrias: On the other hand, in acute/hepatic porphyrias, the exact

molecular basis of the neurological effects is yet to be fully understood. The leading

hypothesis is that the neurological manifestations observed may be as a result of

neurotoxic effects of the high concentrations of circulating aminolevulinic acid (ALA)

and porphobilinogen (PBG) originating from the liver. Alternatively the neurotoxic

effects may be due to deficiency of heme in the neurons 4.

Introduction

8

Figure 3: Pathogenesis of cutaneous effects in porphyria

Acquired Porphyria: Porphyrias can also be triggered by unknown exogenous factors in

the environment such as alcohol, diet, or by a group of drugs called porphyrinogenic

drugs. A number of online databases show long lists of drugs belonging to this latter

class. Commonly, they are drugs that are metabolised mainly by the cytochrome P450

enzymes in the liver. Induction of porphyric conditions by such drugs may be as a result

of increased heme production required for the formation of the cytochrome P450

enzymes via enzyme induction pathways in the liver.

Among other online resources, the website http://www.drugs-porphyria.org has classified

about 1,000 drugs based on their risks of precipitating porphyric attacks 5. Some drugs

http://www.drugs-porphyria.org/

Introduction

9

are to be avoided completely while some may be used if the benefits appear to outweigh

the risks. For example, according to this classification, phenobarbital is porphyrinogenic

while methadone is classified to be ‘possibly porphyrinogenic’ and should only be used

when there is no safer alternative. The hepatitis C virus infection could also be a trigger

for the hepatic dysfunction in porphyria cutanea tarda 6. In addition, a retrospective

population based study in Sweden recommends caution in the use of oral hormonal

contraceptives among women with acute intermittent porphyria 7.

Epidemiology: Porphyria is a rare genetic disease with incomplete penetrance in the

population. This means that some individuals do not show clinical symptoms even

though they carry the genetic defect. Because of the poor penetrance into the population,

the exact prevalence is not known. The prevalence has been found to vary among

porphyria types and also from one specific population to population. The estimated

prevalence can vary from 1 in 500 to 1 in 50,000 people worldwide 8.

1.5 Diagnosis of Porphyrias

Clinical diagnosis of porphyria is frequently challenging firstly because porphyria is a

rare disease and secondly because of overlapping signs and symptoms which may be

easily mistaken for other medical conditions. The accurate clinical diagnosis depends on

detection of specific porphyrin molecules in the body in addition to identifying the

respective clinical signs and symptoms 9.

Detection of excess porphyrins and other intermediates of heme synthesis in urine,

plasma and feces is the first-line diagnostic test. It is important to choose the appropriate

test for each porphyria subtype. For acute porphyrias with acute neurological

Introduction

10

manifestations, detection of urinary δ-aminolevulinic acid and porphobilinogen is the

most sensitive and specific test. On the other hand, determination of total plasma

porphyrins is recommended for cutaneous porphyrias along with skin photosensitivity

reactions. After the initial determination, other specific tests can be done to identify the

specific porphyria subtype present 10.

Although porphyrin molecules may not be significantly increased in asymptomatic family

members of individuals with a confirmed porphyria diagnosis, testing is still generally

recommended since it is a genetic disease. Moreover, since genes responsible for each

porphyria type have been identified, genetic testing in symptomatic patients and further

screening of asymptomatic family members is a promising approach to a more accurate

diagnosis 11.

1.6 Current clinical trials

There are a few recent and ongoing clinical trials in search for potential new therapies for

porphyria. In Europe, a phase III clinical trial of the use of Scenesse® (afamelanotide by

Clinuvel Pharmaceuticals) in Erythropoietic Protoporphyria (EPP) has been recently

completed. In the clinical study, Scenesse® was found to significantly reduce the severity

of pain associated with phototoxic reactions in EPP 12. In February 2012, Clinuvel

submitted an application to the European Medicines Agency (EMA) for Scenesse® to be

used for prophylactic treatment of EPP in Europe 13. Afamelanotide is a synthetic

analogue of α-melanocyte stimulating hormone which works by stimulating the synthesis

of melanin in the skin thus protecting against the damaging effect of the sunlight 14, 15. In

the US, the phase III clinical trial of Scenesse® has also been approved by the FDA; and

the study will be ongoing 16. Also in the US, the use of hydroxychloroquine along with

Introduction

11

phlebotomy is being currently investigated in patients with porphyria cutanea tarda in a

phase II clinical trial 17. Another drug that is currently in phase II clinical trial in France is

deferasirox, an iron chelator. It is believed that forming water soluble chelates with

excessive iron in the body can reduce the production of heme and the accumulation of the

toxic intermediates (i.e. porphyrins) thus relieving the symptoms of porphyria cutanea

tarda (PCT) 18.

1.7 Treatment options and mechanisms:

Treatment of porphyria is directed at specific symptoms of each disease class and will be

discussed under the broad classifications: acute and cutaneous porphyrias.

1.7.1 Acute porphyrias:

For acute porphyrias, drugs that may trigger acute attacks are to be avoided; and

treatment is initiated promptly depending on the prevailing symptoms. For example,

opioid analgesics may be required for the severe abdominal pain. Administration of

intravenous glucose inhibits gluconeogenesis in the liver while indirectly inhibiting the

ALA synthase enzyme required for heme biosynthesis also in the liver by unknown

mechanisms. When administering glucose, it has been recommended that intravenous

glucose should only be used for mild attacks 19. Early administration of intravenous

hemin in form of heme arginate is a safe and effective treatment option 20. Hemin acts by

inhibiting ALA synthase-1 (at the rate-limiting biosynthetic step) through a negative

feedback mechanism in order to prevent further biosynthesis of heme, thereby reducing

the accumulation of both ALA and PBG 21. This prevents further progression of

neuropathic effects. However, hemin may not reverse existing neuropathic conditions,

Introduction

12

and prolonged administration may lead to unwanted results such as iron overload. In

severe iron overload cases, liver transplantation may be required in these patients.

1.7.2 Cutaneous porphyrias:

In cutaneous porphyrias, the skin photosensitivity symptoms can be generally lessened by

avoiding exposure to sunlight. In erythropoietic protoporphyria, antioxidants such as β-

carotene and cysteine are administered orally to improve tolerance to sunlight. β-carotene

exerts its photoprotective effects by quenching reactive species (e.g. singlet oxygen)

generated from photoexcitation of porphyrins on exposure of the skin to sunlight 22.

These antioxidant agents seem to be effective only for erythropoietic protoporphyria

(EPP). Splenectomy or bone marrow transplantation may also be an option.

Lastly, porphyria cutanea tarda (PCT), which is the most common porphyria type is

treated by repeated phlebotomy to reduce iron overload in the liver 23 or by administration

of low dose chloroquine (125mg chloroquine phosphate twice weekly) 24.

Introduction

13

2.0 Literature review

2.1 In vivo localization and excretion of porphyrins:

The physicochemical properties of porphyrin molecules determine their in vivo

localization, the tissue damages they incur and their elimination routes from the body.

The more hydrophilic porphyrins may exist free in the blood circulation while

hydrophobic porphyrins are mostly bound to tissues and blood proteins. Porphyrins have

been shown to bind with various proteins such as human serum albumin, bovine serum

albumin, hemopexin, lipoproteins and globulin among others 25, 26. These proteins may

serve as carriers that transport the hydrophobic porphyrins into the liver where they enter

the enterohepatic circulation 26.

The most clinically significant porphyrin molecules are uroporphyrinogen III,

coproporphyrinogen III and protoporphyrin IX and the detection of their levels in

biological fluids and feces is used in clinical diagnosis 27. In terms of water solubility,

uroporphyrinogen III is the most hydrophilic due to the presence of eight carboxylic

groups on the side chains of its tetrapyrrole ring. Protoporphyrin IX (PPIX) has the least

water solubility with only two carboxylic groups while coproporphyrinogen III with four

carboxylic groups has intermediate water solubility. The solubility properties of

porphyrins have also been correlated with the locations of the cellular damage that they

cause. For instance, the water-soluble uroporphyrinogen III, exerts damaging effects on

hydrophilic cytosol components whereas the lipid-soluble protoporphyrin IX exerts its

damaging effects more on lipophilic targets 28.

Hydrophilic (polar) porphyrin molecules are easily eliminated from the body in the urine

because they are largely ionized and not reabsorbed in the renal tubules during urinary

Introduction

14

excretion. However, for the hydrophobic porphyrins, due to their limited degree of

ionization, they get reabsorbed into the systemic circulation as they pass through the renal

tubules. Hence their major means of elimination will be through secretion into the bile

via the enterohepatic route thus appearing in the feces. For example, protoporphyrin IX is

mainly found in the feces. Conversely, uroporphyrinogen III is largely excreted in the

urine while coproporphyrinogen III has been detected in both urine and feces 29.

2.2 Chloroquine use in porphyria therapy:

Chloroquine (a 4-aminoquinoline derivative; Figure 5), is most widely known for its use

in the treatment and prevention of malaria. Furthermore, chloroquine has found use in a

range of photosensitive skin conditions such as systemic lupus erythematosus 30. Since

porphyria cutanea tarda (PCT) also presents with similar skin sensitivity on exposure to

light, London (1957) decided to try chloroquine in his patient. This was the first

successful administration of chloroquine in PCT resulting in complete clearance of the

skin lesions 31.

Subsequently, chloroquine was adopted for use in other PCT patients albeit with caution.

This was because at high doses, the therapeutic effect of chloroquine was mostly

accompanied by side effects such as fever, nausea, malaise and transient hepatotoxicity

manifesting as abnormal liver enzymes function 32. In 1969, Saltzer and Redecker first

suggested a lower dose of 500mg twice weekly and observed significantly reduced

adverse effects in their patients with this dose 33. However, the potential hepatotoxicity of

the drug at large doses was still a major concern for its use in porphyria. In 1984, Ashton

recommended a low dose of 125mg twice weekly 24, which became widely accepted in

clinical practice up to the present.

Introduction

15

2.2.1 Chloroquine mechanisms of action in porphyria:

Chloroquine (CQ) is thought to relieve the skin photosensitivity in porphyria patients due

to its ability to form unique complexes with the porphyrins and thus possibly increase

their excretion from the body 34. The use of chloroquine has raised issues about its

potential hepatotoxicity, however some scientists have argued that administration of a

low dose minimizes the possibility of such toxicity 24. In its major related application, the

actions of chloroquine as an antimalarial agent have been extensively studied and linked

to its binding with hemin 35. Hemin is a product of hemoglobin digestion, and it is an iron

(III)-containing porphyrin, namely PPIX. Chloroquine and the closely related quinine

molecules can bind with uroporphyrin-I 36. Since chloroquine has been shown to bind to

iron-containing porphyrin moieties, it is not surprising that it may also bind with the iron-

free tetrapyrroles.

Although the initial reports on the effects of CQ in PCT patients have been published

since the 1950’s, the first attempt to describe the molecular basis of its mechanism of

action was not made until 1973 by Scholnick et. al. in the USA 34. Using rats made

porphyric by 3,5-dicarbethoxy-1,4-dihydro-2,4,6-trimethylpyridine (DDC) as an animal

model, an increase in porphyrin content of feces and urine concomitant with a decrease in

the porphyrin content of the liver was observed after administration of chloroquine to the

porphyric rats. This response is similar to what is commonly observed in human clinical

porphyria cases.

In this study, the results of further biochemical experiments concerning the chloroquine

mechanism (leading to porphyrinuria) did not show any increase in δ-ALA synthase nor

any histologic changes in the porphyric rats’ liver compared with the control group. This

Introduction

16

may indicate that the pophyrinuria observed was neither due to increased porphyrin

production nor due to obvious hepatotoxic damage to the liver tissues. In order to probe

further into the cause of the observed porphyrinuria, difference absorption spectroscopy

was used to establish complex formation between some porphyrin molecules and

chloroquine. Also, in vitro dialysis studies showed that chloroquine was able to induce

porphyrin release from liver homogenates saturated with porphyrins. Therefore the

authors proposed the conclusion that complex formation between chloroquine and

porphyrin molecules may be responsible for the porphyrinuria observed both in the

experimental animal models and in human clinical cases.

In another attempt to determine the influence of chloroquine administration on porphyrin

disposition, Goerz et al. employed rats made porphyric by hexachlorobenzene (HCB) as

an animal model 37. This work obtained contrasting results compared with the previous

research by Scholnick et al. The disparity might be as a result of differences in the

chemicals used to induce porphyria in the animal models (DDC and HCB). Goerz et al.

did not find a significant increase in the excretion of porphyrins but found a reduction in

the activity of δ-ALA synthase after chloroquine administration. They therefore proposed

an alternate explanation that chloroquine may act by reducing the activity of δ-ALA

synthase, which is the key enzyme in the heme biosynthetic pathway. The limitation of

both studies could be the use of different animal models, which may not mimic the

pathologic conditions of human porphyria diseases in the same manner. With the

discovery of specific genetic components responsible for each porphyria type, it is

possible to use transgenic animal models in which the respective genes are altered

appropriately to mimic human clinical cases 38.

Introduction

17

2.2.2 Chloroquine –antimalarial mechanism of action

The complexation reactions of chloroquine with ferro- and ferriporphyrin molecules

(heme and hemin) with respect to its antimalarial mechanism of action have been widely

studied. For instance, in 1980, the complexation of quinoline antimalarials (chloroquine,

mefloquine, quinine, quinacrine) with iron (III) porphyrin (hemin) was observed using

difference spectroscopy 39. The stoichiometry of binding for each drug was also

determined by Job’s method. Finally, the authors suggested that the antimalarial action of

these drugs may be through molecular complexation with hemin (a digestive product of

heme).

Later, Sullivan and co-workers (1996) demonstrated that chloroquine exerts its

antimalarial action by first forming a complex with heme. The resulting complex then

binds to hemzoin (a non-toxic heme polymer) thus preventing further polymerization of

heme 35. High levels of toxic heme in the parasites can then lead to their death and thus

cause relief from malaria symptoms in patients. This observation is interesting because

the only difference between the clinical application in malaria and porphyria is that as an

antimalarial agent, chloroquine binds with iron-containing porphyrin while in porphyria,

chloroquine is believed to bind with non-iron heme precursors.

2.3 Quantitative characterization of chloroquine-porphyrin complexes

2.3.1 Aqueous models:

Despite the potential clinical applications of chloroquine-porphyrin complexes in the

treatment of porphyria, there are very few reports on attempts to quantitatively

characterize the complex formation. Moreover, the few reports available are limited to

studies carried out in aqueous buffered medium. While this is fairly easy for most water

Introduction

18

soluble porphyrin molecules, it is very difficult for the study of protoporphyrin IX due to

its poor water solubility.

In one of the earliest reports in the literature, the stoichiometry and strength of

chloroquine-porphyrin complexation was examined by optical absorption difference

spectroscopy40 in aqueous systems. Specifically, the research studied the complexation of

chloroquine with coproporphyrin, uroporphyrin and hematoporphyrin in phosphate

buffered saline (pH7.4) environment. Other studies by Constantinidis and Satterlee

examined the binding of quinine and chloroquine with uroporphyrin I and urohemin I,

also in aqueous systems, using a combination of UV-visible spectroscopic methods to

calculate the stoichiometry and association binding constants 36, 41. In addition, by using

13C-NMR spectroscopy, the geometric structures of the complexes were examined.

For quinine, the presence of the 9-hydroxy group on its quinoline ring was shown to give

a facilitated pathway for coordination with the iron atom present on the urohemin

structure. This coordination reaction was found to be in addition to a rather limited π-π

interaction between the tetrapyrrole ring and the quinoline ring 41. However, in the case of

chloroquine binding with uroporphyrin I, the structure was proposed to be a cofacial π-π

interaction between the ring systems. The authors then proposed that a weaker

association constant obtained for quinine-uroporphyrin I complexation compared with

that of choloroquine may be due to the limited π-π bonding found between quinine and

uroporphyrin I 36.

Introduction

19

2.3.2 Need for hydrophobic models:

Generally, the major limitation of studying porphyrin complexation reactions in aqueous

models is the fact that hydrophobic porphyrin molecules tend to self-aggregate in the

process of hiding their hydrophobic rings from the aqueous environment. This widely

reported observation severely limits the use of aqueous studies for the study of

hydrophobic porphyrin complexation with other molecules. For instance, at aqueous

concentrations below 4μM, porphyrins such as PPIX exist predominantly as dimers while

formation of more complex aggregates occurred at 4μm 42. When dimerization occurs, the

Beer- Lambert plot gives a non-linear relation of absorbance versus concentration for the

monomers, but isosbestic points will be evident in the absorption spectra of the

monomers and dimers that are in equilibrium with each other. However, hydrophilic

porphyrin molecules in the aqueous system usually stay in their monomeric forms and

give a linear absorbance-concentration plot. It is believed that this aggregation behaviour

of hydrophobic porphyrin molecules is not favoured in organic solvents 43, at least within

the concentration range employed in this study.

Protoporphyrin IX (PPIX), which has poor water solubility, may also be solubilized by

adjusting the pH by treatment with strong acid and base. However, this process may lead

to some degradation of the porphyrin ring. Hence, it is more logical to study the

complexation of hydrophobic porphyrin molecules such as PPIX in non-aqueous systems,

where they would be readily soluble. Furthermore, chloroquine itself also has the ability

to self-associate in aqueous solutions. 1H-NMR studies of the dimer complex have shown

them to be closely stacked rings, slightly dephased with π-π hydrophobic interactions

between the ring systems 44. Our motivation to study porphyrin complexation reactions in

Introduction

20

hydrophobic models is therefore twofold: firstly to avoid the self-aggregation of PPIX

that will occur in an aqueous medium and secondly to ensure proper dissolution of PPIX.

Moreover, a hydrophobic model will better mimic the in vivo environments such as lipid

membranes where the hydrophobic porphyrin molecules may be located 45 46.

2.4 Protoporphyrin IX molecule:

Protoporphyrin (PPIX) is one of the major clinically significant porphyrin molecules in

humans 27. It is therefore the focus of this research. PPIX is the last intermediate in the

biosynthesis of heme (Figure 2), and ferrochelatase is the key enzyme responsible for

converting protoporphyrin IX into heme. The deficiency of the ferrochelatase enzyme

results in the accumulation of PPIX in the blood and hence the manifestation of

erythropoietic protoporphyria (EPP). Because of the very hydrophobic nature of PPIX

and its molecular weight (>500g), it is only eliminated from the body through the

enterohepatic route (feces) and not through the urine 2. Therefore, when PPIX

accumulates in the body due to ferrochelatase deficiency as in erythropoietic

protoporphyria (EPP), its elimination may be a challenge compared to other more water

soluble porphyrins.

Furthermore, unbound PPIX does not exist in the blood under normal conditions.

However in the case of EPP disease state, PPIX is mostly bound to hepatic tissues and to

proteins, particularly the albumins in the blood 47. It is believed that PPIX binding to

proteins such as albumin and hemopexin plays a role in their clearance from the body.

That is, these proteins may serve as carriers that transport the PPIX molecules into the

liver to enter the enterohepatic circulation 26. In addition, formation of complexes with

PPIX may enhance its elimination from the body through the enterohepatic route.

Introduction

21

Because of their photochemical properties, specific porphyrin molecules (such as

hematoporphyrin) have also found use as photosensitizers in photodynamic therapy in the

treatment of malignant tumors 48. However, their complex photodynamic mechanisms are

not the focus of this research.

2.5 Molecular interactions: binding and complexes

One of the most important issues that some medicinal chemists and pharmacologists are

interested in studying is the nature of interactions between drug molecules and potential

targets in the body. This information may help shape better understanding of the drug’s

mechanisms of action and may lead to the development of new therapeutic molecules for

the disease condition. For instance, knowledge about the formation and strength of

complexation reactions between PPIX and a range of acceptor molecules may be useful

in porphyria therapy.

The interaction of molecules to form complexes may involve a number of intermolecular

binding interactions such as hydrogen bonding, π–π bonding through van der Waals

forces of attraction, and electrostatic attraction between positively charged and negatively

charged centres. The latter attraction generally is the strongest binding force between

two molecules. Figure 4 in particular shows a theoretical superimposition of PPIX and

CQ molecules in a 1:1 π to π complex. π to π bonding is a relatively weak force of

attraction (compared with covalent bonding or electrostatic bonding) that exists between

aromatic compounds due to induced dipolar interactions between the polarizable π-

systems on the respective aromatic rings. Charge transfer complexes in π to π complexes

involving electron donor–electron acceptor systems have also been studied among a wide

range of compounds 49. However, electronic charge transfer does not always occur

Introduction

22

between the two aromatic molecules in the van der Waals interaction complexes; and

frequently the induced dipolar interactions are predominant. Both the induced dipolar

complexes and the charge transfer complexes (if formed) can be easily detected using

optical absorption difference spectroscopy. For example, Gonzalez has studied complex

formation between benzoquinones and a tetraphenylporphyrin derivative using optical

absorption difference spectroscopy and fluorescence measurements 50.


a 1:1 complex

Apart from the electronic donor-acceptor interactions described above, it is also possible

to have a proton donor-acceptor interaction between the PPIX and the acceptor molecule.

For instance, the carboxylic acid on PPIX can release a proton to the basic nitrogen on

chloroquine so that the carboxylic acid becomes positively charged while the basic amine

Introduction

23

becomes protonated as indicated in Figure 4. This form of electrostatic interaction may

provide an additional force of attraction for the intermolecular binding interactions.

The affinity of binding between two given molecules will depend on the combination of

forces holding them together. Depending on the affinity of binding, different

mathematical models have been established for the quantitative determination of the

equilibrium constants. For instance, the Benesi–Hildebrand plot and its various

modifications have been employed in the measurements of binding constants for weak

complexes 50. In the case of stronger interactions, the dissociation constants can be

deduced directly from the binding curve. In addition, the nature of the strong complexes

can be analysed further using the Scatchard and Hill equations and associated plots.

Introduction

24

3.0 Other compounds tested:

Figure 5 shows the structures of the acceptor molecules tested for complexation with

PPIX in this research.

3.1 Chloroquine related compounds:

Amodiaquine is a 4-aminoquinoline antimalarial that is similar to chloroquine in action. It

is a very pale yellow odorless powder with pKa values of 7.1 and 8.1. The dichloride

dihydrate salt form is soluble in water 51.

Mefloquine is a 4-hydroxymethyl-quinoline antimalarial that is structurally related to

quinine. It is used for the treatment of chloroquine-resistant malaria and also for malaria

prophylaxis. It is a white or slightly yellow crystalline powder with pKa value of 8.5. The

hydrochloride salt form is very slightly soluble in water, soluble in alcohol and freely

soluble in methyl alcohol 51.

Pyronaridine is a naphthyridine derivative used in China in the treatment of vivax

malaria and chloroquine-resistant malaria 51. It is a relatively new antimalarial agent with

promising use when combined with artesunate 52. It is a yellowish crystalline powder with

pKa values of 10.30, 9.88, 7.39, and 7.08. The tetraphosphate salt is soluble in water.

Quinacrine is a 9-aminoacridine antiprotozoal used as an alternative to nitroimidazole in

the treatment of giardiasis 51. It is a bright yellow crystalline odorless powder with bitter

taste and pKa values of 7.7 and 10.3. The hydrochloride salt is soluble in water, alcohol

and almost insoluble in chloroform 53.

Introduction

25

Quinine is a 4-hydroxymethylquinoline antimalarial agent obtained from the bark of the

cinchona tree. It is reserved for the treatment of chloroquine resistant malaria. It is a

white, fine, needle-like crystals with pKa values of 8.1 and 10.1. The hydrochloride salt is

soluble in water and freely soluble in alcohol 51.

3.2 Simple ring systems:

Duroquinone is a 2,3,5,6-tetramethyl substituted 1,4-benzoquinone. It is not clinically

used as a drug. It is a dark yellow powder soluble in acetone.

Caffeine is a xanthine derivative obtained from natural origin. It is probably the most

widely used central nervous system stimulant, and it is used therapeutically. Caffeine is a

white odorless powder with bitter taste. It is an extremely weak base and so does not

form stable salts 53. It is sparingly soluble in water and soluble in boiling water 51.

3.3 Other ring systems

Amitriptyline is a dibenzocycloheptadiene, a tricyclic antidepressant developed from the

phenothiazine and structurally related to chlorpromazine. It is used clinically in the

treatment of depression. The hydrochloride salt is a white powder, freely soluble in water

alcohol and dichloromethane 51.

Chlorpromazine is an older phenothiazine antipsychotic agent. It is used in the treatment

psychosis and chemotherapy-induced nausea. The hydrochloride salt is a white

crystalline powder, very soluble in water and freely soluble in alcohol 51.

Introduction

26

Figure 5: Structures of electron acceptor molecules under study.

Introduction

27

4.0 Experimental design and approach to be employed:

Optical absorption difference spectroscopy in hydrophobic models was employed to

study complex formation between PPIX and the selected acceptor molecules. The optical

absorption difference spectroscopic method thus involves the measurement of changes in

the absorption spectrum of PPIX upon complexation.

Normally, PPIX spectrum is made up of five bands. The first band, which occurs at

400nm, is very intense and it is called the Soret band. The other four bands, known as the

Q-bands, occur between 500nm and 700nm (Figure 6). If complex formation occurs, we

expect changes in the spectrum of PPIX after addition of acceptor molecules. These

spectral changes appear as a broadening of the spectrum resulting in smaller peaks and

less shallow troughs (Figure 7). This produces negative and positive changes in

absorbance values at the peaks and the troughs, respectively. These observed changes can

be plotted on their own to give the difference spectra.

Figure 6: Spectrum of protoporphyrin IX (36μM) solution in acetone showing the four

Q-bands

-0.1

6E-16

0.1

0.2

0.3

0.4

0.5

0.6

480 530 580 630 680

Ab

sorb

ance

Wavelength(nm)

PPIX Spectrum (36μM)

Introduction

28

Figure 7: Spectral changes as expected in the protoporphyrin IX spectrum upon

complexation

The measurement of all of the difference spectra is based on the Beer-Lambert Law. The

Beer-Lambert Law defined in Figure 8 represents the fundamental relationship between

absorbance and the concentration of the absorbing species in the absorption spectrum at

any given wavelength.

Wavelength

Absorbance

nce

ΔA

ΔA

480nm 600nm

Introduction

29

Figure 8: Illustration of the Beer-Lambert Law

When absorbance is plotted against the concentration, according to the Beer-Lambert

law, a straight line is obtained at low absorbance values. Generally, when the absorbance

value is higher than 1, the Beer-Lambert relation becomes non-linear and unreliable due

to inhomogeneous light absorption in the optical path length. Thus, caution should always

be taken into consideration when studying molecules at higher absorbance values.

Introduction

30

4.1 Hydrophobic models:

The hydrophobic models used in this study are: homogenous organic solvent mixtures

such as 50:50 acetone /dichloromethane, Triton X-100® (Figure 9) in an aqueous

micellar system and an aqueous albumin system that binds PPIX. These models are

important in order to avoid self-association of PPIX that occurs at sufficiently high

concentrations in aqueous environments and also because most of the selected

compounds are hydrophobic. In addition, hydrophobic models better mimic the potential

biological location of PPIX in the body such as in the stratum corneum of the skin.

Acetone/dichloromethane 50:50 solvent system:

The 50:50 acetone/dichloromethane homogenous solvent system was chosen because it

affords optimum solubility of the molecules under test. PPIX is more soluble in acetone

while chloroquine and similar structures in the free base form are more soluble in

dichloromethane.

Triton X-100® aqueous micellar systems

Figure 9: Triton X-100® structures

Introduction

31

Triton X-100® is a nonionic surfactant that has the ability to self-aggregate when

dispersed in oil or water due to its amphiphilic nature. When dissolved in water, the

molecules assemble so as to hide their hydrophobic portions in the core while exposing

their hydrophilic portion to the aqueous environment. The resulting aggregates are known

as micelles. In the Triton X-100® aqueous micellar systems, PPIX and the acceptor

molecules tested were dissolved in the hydrophobic core of the oil in water micelles. The

micellar systems at appropriate concentrations have been used as in vitro models of lipid

membranes 54. The oily core of the micelles will be sufficiently hydrophobic to drive the

complexation reaction.

Bovine Serum Albumin Systems

Serum albumin is the most abundant protein found in the blood. One of the most striking

properties of albumin is its ability to reversibly bind a wide variety of ligands. Thus

albumin plays a principal role in the transport, metabolism and distribution of both

endogenous and exogenous ligands 55. Bovine serum albumin (BSA) is a serum albumin

protein obtained from cows. BSA is often used in various biochemical experiments.

In vivo binding of porphyrins with albumin has been observed in the blood 26. Therefore

aqueous albumin systems may be a useful in vitro model of studying porphyrin

complexation. Albumin has multiple binding sites. In this model, PPIX is expected to

bind to the hydrophobic sulfhydryl binding sites on albumin.

For all of the hydrophobic models, the absorption spectrum changes profiled for PPIX in

Figure 7 are very relevant where the occurrence of the characteristic positive and

negative changes in absorbance (ΔA) is an indication of probable complex formation

Introduction

32

between PPIX and an acceptor molecule in these systems. To confirm complex

formation, the PPIX spectrum were measured and analyzed as a function of increasing

acceptor concentration. These experiments were carried out in triplicate measurements at

each acceptor concentration, and the average ΔA values obtained were used to deduce the

affinity of binding with the use of already established mathematical and graphical models

to be described in the next section. For instance, for weak complexes, the equilibrium

constants were determined using an appropriate modification of the Benesi-Hildebrand

equation 49. In addition, Hill and Scatchard plots were used to measure the apparent

association constants and the stoichiometry of binding for stronger complexes 41.

4.2 Graphical and Mathematical models:

The Benesi-Hildebrand (B-H) plot for weak complexes:

The Benesi-Hildebrand (B-H) plot is used in physical chemistry to determine association

constants of complexes involving weak non-bonding interactions. When the B-H plot and

its derivatives are to be used, one of the complexing molecules must be in excess by at

least 10 fold. In this experiment, the concentration of PPIX (36µM) was kept constant,

while the concentrations of the electron acceptor molecules were varying and in excess.

Where [Q] = Molar concentration of acceptor molecule

[P] = Molar concentration of protoporphyrin IX

Δε = Change in absorptivity of protoporphyrin IX

............Equation 1

Introduction

33

Equation 1 above is a modification of the B-H equation 49. When

is plotted

against , the resulting plot is expected to be a straight line. The association constant

(Ka) can then be easily determined from the values of the slope and intercept.

From the relation in Equation 1,

Hill and Scatchard plots for strong complexes

When binding occurs at a sufficiently low concentration of the acceptor molecules, this

may be an indication for stronger binding interactions. The binding constants may then be

deduced directly from the binding curve (plotting ΔA against electron acceptor

concentration). The dissociation constant may be obtained from the binding curve – that

is the concentration that corresponds to 50% of maximum binding (ΔAmax).

Hill and Scatchard plots can also be used to further analyse the binding curve data

particularly when multiple binding sites are involved 56.

[

] ............Equation 2

Equation 2 above is known as the Hill equation, named after the formulator. In the Hill

plot, [

] is usually plotted against giving a straight line. The plot is used

to determine cooperativity among multiple binding sites. In addition, the Hill plot also

gives an indication of the stoichiometry of binding. For instance, for a 1:1 complex, the

Introduction

34

slope n (also called the Hill parameter) is expected to be equal to 1 and the dissociation

constant (Kd) may be read from the graph.

Furthermore, the Scatchard plot can be plotted following the Scatchard equation

(Equation 3).

..............Equation 3

When

is plotted against , the nature of the resulting curve may be used to explain the

nature of binding. When a straight line is obtained on the Scatchard plot, this may

indicate involvement of a single/independent binding site. The dissociation constant can

be read from the plot following the relation below.

or

However, a non-linear behaviour on the Scatchard plot may indicate interactions among

multiple binding sites. Although this model was designed for binding of proteins or drugs

at their receptors, in the context of complex formation, if we observe this form of non-

linear behaviour, it may indicate formation of more than one structural type of complex

due to conformational degrees of freedom. Each of these unique complexes would form

due to locking in of specific conformations of the PPIX pendant (propionic acid)

carboxylic acid groups and also due to locking in specific conformations on the various

Introduction

35

acceptor molecules under study. It is presumed that specific conformations in both

molecules are required for the formation of high affinity complexes.

Rationale and Hypothesis

36

CHAPTER II: RATIONALE AND HYPOTHESIS

1.0 Rationale for the research:

In contrast to its antimalarial mechanisms, the mechanism of actions of chloroquine in

porphyria is highly understudied. The few studies that were done in the past were limited

to hydrophilic systems. There is, therefore, a need to carry out studies in hydrophobic

models that mimic in vivo membrane systems in order to understand chloroquine’s

actions in quenching photosensitive skin reactions in porphyria patients.

Chloroquine is used in the treatment of PCT. It is believed that chloroquine relieves the

photosensitivity reactions in porphyria patients through formation of water soluble

complexes with excess porphyrins in the liver while enhancing their urinary excretion 34.

Although this observation has existed in the literature since the 1970’s, there are limited

reports on the molecular interactions between chloroquine and the porphyrins. Most of

these few previous works are limited to the study of water soluble porphyrin-chloroquine

complexes in aqueous systems 36, 40, 41. There are no reports on reactions of poorly water

soluble porphyrins such as PPIX in hydrophobic systems, which mimic more closely the

lipid in vivo tissue environments such as in the skin. However, the complexation

reactions of chloroquine with heme and hemin have been examined in a greater detail

with respect to its application as an antimalarial agent 35, 36. Thus, the exact nature of the

chloroquine-porphyrin complexes in porphyria patients is yet to be fully understood.

Deeper insights into the nature and properties of these complexes in both aqueous and

non-aqueous environments may lead to the discovery of new molecules that can be useful

in porphyria therapy.


37

The focus of this project is on protoporphyrin IX (PPIX). Other porphyrin molecules may

also be studied in a similar manner. Chloroquine is believed to form π-π complexes with

porphyrins such as PPIX in a system where chloroquine is the potential electron acceptor

and PPIX is the potential electron donor. While the electron transfer mechanism is

unproven in these complexes, it is clear that induced dipolar interactions of the π-

electrons will provide a significant van der Waals attraction between the two π-electron

systems when they are in cofacial proximity to each other. Based on this proposition, this

research intends to explore some other molecules similar to chloroquine that may bind to

PPIX in a similar manner. Ultimately, this mechanism of action may be exploited in the

design of better alternatives for porphyria treatment in the future.


38

1.1 Hypotheses of the Research

We hypothesize that:

A hydrophobic environment (in vitro) may favour complex formation between

porphyrin(s) such as PPIX and acceptor molecules.

It may be possible to determine the binding affinities of these complexes (in vitro)

as an indication of each molecule’s potential use in porphyria therapy.

1.2 Objectives of the Research:

From the outcomes of this research, we were able to:

Establish complex formation between PPIX and a range of acceptor molecules in

selected hydrophobic models in vitro.

Measure the dissociation constant Kd (as a measure of the binding affinity) for the

complexation of PPIX with each electron acceptor using appropriate

mathematical models.

Identify the specific structural features that are important for complex formation.

Materials and Methods

39

CHAPTER III: MATERIALS AND METHODS

1.0 Materials

Table 2: Table of chemicals

MATERIALS Sources

Acetone Sigma Aldrich, Acros

Amitriptyline hydrochloride Sigma Aldrich

Amodiaquine dichloride dihydrate Sigma Aldrich

Caffeine Sigma Aldrich

Chloroquine diphosphate salt Sigma Aldrich

Chlorpromazine hydrochloride Sigma Aldrich

Dichloromethane (spec. grade >99.5%) Sigma Aldrich, Fisher

Duroquinone 97% Sigma Aldrich

Mefloquine hydrochloride Sigma Aldrich

Protoporphyrin IX base (95%) Sigma Aldrich

Pyronaridine tetraphosphate Sigma Aldrich

Quinacrine dihydrochloride Sigma Aldrich

Quinine hydrochloride dihydrate Sigma Aldrich

Sodium carbonate (reagent grade 99.5) Fisher scientific

Triton X-100 Sigma Aldrich

Trizma base (Tris) (reagent grade 99.9%) Sigma Aldrich

Trizma HCl (Tris HCl) buffer 1M solution

pH7.40

Fluka analytical

Trizma hydrochloride (Tris HCl)

(reagent grade >=99.0%)

Sigma Aldrich


40

2.0 Acetone: Dichloromethane 50:50 solvent system

2.1 Preparation of Stock solutions

72.3 µM Protoporphyrin IX solution

8.140 mg of protoporphyrin IX (PPIX) base crystals [Sigma-Aldrich, F.W. 562.66g] was

weighed using a [Fisher Scientific] microbalance. Because of the hard nature of the PPIX

crystals and in order to allow for complete dissolution, the crystals were first crushed in a

small quantity of acetone in a 50 mL glass round bottom flask. The initial small volume

was transferred into a 200 mL volumetric flask using a pipette. The round bottom flask

was rinsed with acetone and made up to volume in the 200 mL volumetric flask.

Duroquinone Solutions:

1.78 g of duroquinone [Sigma-Aldrich, F.W. 164.2g] was weighed and dissolved in 25

mL of acetone [Acros; spectrophotometric grade] to make a 0.4335 M duroquinone stock

solution.

Quinine solutions:

0.792 g of quinine hydrochloride dihydrate [Sigma-Aldrich, F.W. 396.92g] was weighed

and dissolved in 20 mL of dichloromethane [Sigma-Aldrich; spectrophotometric grade]

to make a 0.400 M Stock solution.

Caffeine solutions:

0.388 g of caffeine [Sigma-Aldrich, F.W. 194.19g] was weighed and dissolved in 10 mL

dichloromethane to make a 0.200 M stock solution.


41

Amitriptyline solutions:

0.628 g of amitriptyline hydrochloride [Sigma-Aldrich, F.W. 313.87 g] was weighed and

dissolved in 10 mL of dichloromethane to make a 0.200 M stock solution.

Chlorpromazine solutions:

2.13 g of chlorpromazine hydrochloride [Sigma-Aldrich, F.W. 355.33 g] was dissolved in

20 mL dichloromethane to make a 0.300 M stock solution.

2.2 Extraction of free base

For compounds supplied in the salt forms that are insoluble in the dichloromethane

solvent, a simple extraction process was required to prepare the stock solutions. Below

the extraction process for chloroquine is given in some detail. Basically the same process

was followed for amodiaquine, quinacrine, mefloquine and pyronaridine. For mefloquine,

an additional solubilization process was required as indicated below.

Aqueous chloroquine diphosphate salt solution:

0.515 g of chloroquine diphosphate [Sigma-Aldrich, F.W. 515.87g] was dissolved in 10

mL of deionized water in an Erlenmeyer flask to make a 0.100 M aqueous solution.

Dichloromethane phase and neutralization:

10 mL of dichloromethane was then added to the aqueous salt solution from above. 0.265

g of sodium carbonate [Fisher Scientific; reagent grade F.W. 106] was added slowly

while stirring. The pH of the upper aqueous layer (typically 9.33) was measured using a

pH meter (Fisher Scientific). The mixture was allowed to mix overnight using a stir bar to


42

allow for proper partitioning. Over the mixing period, the free base would have

partitioned to the dichloromethane layer.

Separation of the 2 layers:

The two phases: aqueous and dichloromethane layers were separated using a 60 mL

separating funnel. The dichloromethane layer (with higher density) was collected

carefully from the funnel as the lower layer. The extraction process was repeated with 3

more 10 mL aliquots of dichloromethane to ensure complete extraction. The four

dichloromethane extracts were then pooled together.

Suitable dilutions of each basic acceptor molecule combined extract were made to obtain

the desired concentration ranges used in the final solutions where the porphryin (PPIX)

was also present.

To demonstrate complete extraction, the spectrum of the aqueous solution in the spectral

range 300-500 nm was monitored before and after the extraction process to show that

only traces of the molecules were left behind in the aqueous phase. This was done by

taking the spectrum of each molecule in the aqueous medium before and after extraction.

The pH of the aqueous layer was also measured after extraction

Solubilization of mefloquine:

Mefloquine is a poorly water soluble compound. Proper dissolution in water was ensured

by pH adjustment to pH 3.86 with acetate buffer (20 mM) to make a concentration

equivalent to 250mg in 141 mL and allowed to mix for 1 hour as suggested by Strauch S.

et al 57. After complete dissolution was achieved, the extraction process was continued by


43

adding the dichloromethane layer and continuing the four-fold extraction into

dichloromethane as described above.

2.3 Complexation mixtures in homogenous solutions containing PPIX and

each acceptor molecule

Preparation of each complexation sample in the solvent mixture:

Equal volumes (5 mL) of dilutions of stock solutions for each test compound (with

varying concentrations) in dichloromethane and 72.3 μM PPIX in acetone (prepared

above) were mixed together in an Erlenmeyer flask to get a series of complexation

sample mixtures with differing concentrations over an appropriate range for the

determination of the extent of PPIX complexation with each of the acceptor molecules.

Preparation of the PPIX (36.15 μM) reference in the homogenous solvent mixture:

Each PPIX reference solution for the measurement of the difference spectra was prepared

at the same concentration by mixing 5 mL of neat dichloromethane and 5 mL of PPIX in

acetone stock solution (72.3 µM) in an Erlenmeyer flask to obtain a final concentration of

PPIX at 36.15 μM in the 50:50 dichloromethane / acetone homogenous solution.

2.4 UV-Visible difference spectroscopic analysis

The spectra of the PPIX complexation with each basic acceptor test compound were

recorded in the range of 480 – 800 nm but actually examined for data analysis in the

porphyrin Q-band region in the range of 500 – 650 nm using the UV-Visible absorption

spectrophotometer {Thermoscientific; Evolution 60}. The same spectrophotometer was

used for all spectroscopic readings in the experiments, and it was found to give


44

reproducible results. In each case, each complexation sample mixture was placed in the

10 mm or 1mm path length suprasil sample cuvet (Fisher Scientific) while a blank

acetone/dichloromethane 50/50 solvent system was placed in the optically identical

suprasil reference cuvet. The spectrum of the PPIX 36.15 µM reference solution in the

same 50:50 dichloromethane / acetone homogenous solution was recorded in exactly the

same wavelength region.

The spectrum of the PPIX (36.15 µM) reference was then subtracted from the spectrum

of each of the complexation mixtures for all of the PPIX acceptor test compounds. As

mentioned above, the difference in absorption values over the 500 – 650 nm range

represents the difference spectra that were analyzed for each complexation mixture. The

experiments were performed with triplicate measurements, and the average and standard

deviation of delta absorbance values at selected wavelengths were determined for each

difference spectrum in order to present useful binding curves for complexation of PPIX

with each acceptor molecule.

Background absorption subtraction:

For most of the molecules tested (except for chloroquine and caffeine) there was a need

to further subtract the underlying residual background absorption spectra of the

individual acceptor molecules. This is because most of these acceptor molecules have a

significant tailing background absorption in the region under study (500-650 nm). Figure

10 shows the absorption spectrum of PPIX at 36.15 µM overlaid with the background

absorption of the molecules being studied at the highest concentration used in the

experiments. From Figure 10 below, it should be noted that pyronaridine has the greatest


45

absorption in this region. It was necessary to treat the pyronaridine data as a distinct case

in the analysis section.

Figure 10: Plots showing background absorption of acceptor molecules

To plot the difference spectra for each PPIX-acceptor complex, the delta absorbance

values (ΔA values) observed at 1 nm wavelength intervals over the 500-650 nm range

domain were used. All of the spectra were actually recorded over the 480-800 nm range

in order to achieve establish good baselines for the difference spectra. The delta

absorbance data for each spectrum was then used to produce the respective Benesi-

Hildebrand, Hill and Scatchard plots (see sections 5.1 to 5.4 in the following text) and to

make a quantitative determination of the dissociation constants for each of the test

acceptor compounds that formed complexes with PPIX.

It was usually necessary to adjust the difference spectra subtractions so that the Q-bands

of the porphyrin spectrum were completely removed and secondly to achieve an

acceptably flat baseline in the 700-800 nm region. This fine tuning of the subtractions for

each difference spectrum was performed by using Excel spread sheets with variable

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

480 530 580 630 680

Plots showing background absorption

PPIX_36.15 µM

AMQ_10.75 mM

MFQ_3.670 mM

QCR_8.525 mM

PYR_0.916 mM


46

weighting factors for the PPIX reference spectrum for each of the PPIX-acceptor

samples. The main goal was to achieve acceptable difference spectra that could be

superimposed on each other to compare the correct delta absorbance values over the

wavelength range where the difference spectrum was significant and reproducible: e.g.,

from 480 nm to 700 nm.

3.0 Triton X-100® aqueous micellar systems

Preparation of the Tris HCl buffer 62.5 mM:

12.5 mL of 1 M Tris HCl buffer solution (Fluka analytical, pH 7.40) was measured and

diluted to 200 mL with deionized water in a 200 mL volumetric flask to give a

concentration of 62.5 mM. The pH was confirmed after dilution with the pH meter

[Fisher Scientific].

Preparation of protoporphyrin IX (0.0101%w/w) in neat Triton X-100®

3.33 mg of PPIX was weighed on a microbalance [Fisher Scientific, Mettler M5] and

dissolved in 32.883 g of neat Triton X-100® in a 50 mL beaker. The PPIX was left

overnight while stirring with a stir bar to allow for complete dissolution.

Preparation of the Triton X-100® aqueous micellar test samples:

Due to its limited aqueous solubility at pH 7.40, the AMQ was initially dissolved in an

unbuffered environment (actually at pH ~5) whereas chloroquine was studied in Tris HCl

buffer (pH 7.40).

0.5159 g of chloroquine diphosphate [Sigma-Aldrich, F.W. 515.87g] was dissolved in 10

mL of deionized water (pH 6.13) to form a 100 mM chloroquine aqueous stock solution.


47

Secondly, 20 mL of 62.5 mM Tris buffer was added to the 10 mL, 100 mM chloroquine

solution in a 50 mL beaker. 10 g of PPIX in neat Triton X-100® (prepared above) was

then added and dispersed in the buffered chloroquine solution to form micelles. The

initial micellar dispersion was transferred to a 50 mL volumetric flask using a pipette.

The beaker was then rinsed with 62.5 mM Tris buffer and made up to the 50 mL volume

to give a micellar system (pH 5.87) of 20.0 mM chloroquine, 36.1 µM PPIX in Triton X-

100® (20% w/v) dispersion and 20 mM Tris HCl buffer.

0.046 g of amodiaquine dichloride dehydrate [Sigma-Aldrich, F.W. 464.81g] was

dissolved in 3 mL of deionized water (pH 6.13) in a 10 mL beaker. 1 g of PPIX in neat

Triton X-100® (prepared above) was added and then dispersed with water to form oil-in-

water micelles. The initial micellar dispersion was transferred to a 5 mL volumetric flask

using a pipette. The beaker was rinsed with water and made up to 5 mL volume to give a

micellar system (pH 4.5) of 20.0 mM AMQ and 36.1 µM PPIX in the Triton X-100®

(20% w/v) dispersion.

The procedure was repeated to give varying concentrations of each acceptor molecule

(chloroquine or amodiaquine) in each final micellar dispersion for the measurement of

the optical absorption difference spectra in the Triton X-100® (20% w/v) dispersions.

Preparation of the Triton X-100® aqueous micellar reference sample

1g of PPIX in neat Triton X-100® (prepared above) was weighed and dispersed in 3

mL of deionized water (pH 6.13). The initial micellar dispersion was transferred into a 5

mL volumetric flask using a pipette. The beaker was then rinsed with water and made up


48

to the 5 mL volume to give the reference micellar system (pH 5.60) of 36.1 µM PPIX in

Triton X-100® (20% w/v) dispersion.

UV-Visible difference spectroscopic analysis

The spectrum of PPIX in the 20 w/v% Triton X-100® aqueous micellar reference sample

was subtracted from that of the test samples containing both the PPIX (same

concentration) and the acceptor molecules in the same solvent dispersion to give the

difference spectra for complexation for both chloroquine and amodiaquine.

Relatively high concentrations in the millimolar range of chloroquine were required to

obtain a significant difference spectrum in these 20% w/v Triton X-100 aqueous

dispersion experiments. These observations will be explained in the Results section.

4.0 Bovine Serum Albumin (BSA) systems

Preparation of 1% w/v BSA solution: 1 g of BSA was weighed and dissolved in

100 mL of 50mM Tris buffer (pH 7.40) in a volumetric flask to make a 1 %w/v BSA

solution.

Reference preparation: 1g of BSA was weighed and triturated with 2.0256 mg

of PPIX crystals in a mortar. The triturated mixture was then dissolved in 50mM Tris

HCl buffer (pH 7.40) and made up to 100 mL to give 1% BSA and 36.35 µM PPIX. This

was allowed to stay overnight (in the absence of light at room temperature) for proper

dissolution.

Sample preparation: Aliquots of chloroquine (100 mM) aqueous stock solution

were added to the reference BSA/PPIX mixture (prepared above) to give the sample


49

preparations that contained varying concentrations of chloroquine along with the PPIX at

the same concentration (36.35 µM) as in the above reference solution.

Spectroscopic analysis:

The spectrum of PPIX in the presence of BSA was obtained by measuring the

spectrum of the reference preparation (sample cuvet) against the 1% w/v BSA solution

(reference cuvet).

The spectrum of the sample preparations (sample cuvet) were measured against

the reference preparation (reference cuvet) to obtain the difference spectra.

Chloroquine and BSA: 0.1g of BSA was weighed and dissolved in 10mL of

50mM Tris HCl buffer (pH 7.40). An aliquot (4 µL) of chloroquine (100mM) stock

solution was added to give measurable chloroquine concentration (40 µM). The spectrum

of the BSA/ chloroquine mixture was measured against the BSA solution alone. The

expected spectrum of chloroquine was then observed. The procedure was repeated using

a Tris HCl buffer of pH 9.00 to determine the effect of pH and ionization on the

chloroquine spectrum.

5.0 Graphical and Mathematical models and statistical analysis


50

5.1 Benesi-Hildebrand Equation and Plot: plotted for duroquinone, quinine,

and caffeine. The data was analyzed by a weighted least squares averaging method.

5.2 Binding curves: were plotted for chloroquine, amodiaquine, mefloquine,

quinacrine and pyronaridine. Averaged triplicate values of the delta absorbances were

plotted against molar concentrations for each acceptor molecule in the binding curves.

For amodiaquine, separate plots for an n=6 determination were also plotted to make a

comparison with the n=3 determination and plots.

5.3 Hill plots: were plotted for chloroquine, mefloquine, amodiaquine, and

quinacrine as shown. Triplicate measurements of the delta absorbances were used to

generate triplicate Hill plots. Kd was obtained from each Hill plot. Averages and standard

deviations of the triplicate Kd values were computed for each acceptor molecule to give

the values in Table 4 in the results section. For amodiaquine, separate plots for n=6

determination were also plotted.

Scatchard plots: plotted for chloroquine, mefloquine, amodiaquine, and

quinacrine as shown. Averaged triplicate values of the delta absorbances were used to

plot the Scatchard plots as shown in the results section.

Results and Discussion

51

CHAPTER IV: RESULTS AND DISCUSSION

1.0 Acetone: Dichloromethane 50:50 homogenous solvent system

The difference spectra shown in the results were obtained by subtracting the PPIX

reference spectrum from PPIX-acceptor molecule complex spectra. Spectral changes on

the PPIX spectrum occur as a result of complex formation which causes broadening of

the PPIX spectrum (Figure 7). The difference obtained (∆A) is then plotted against the

wavelength for each acceptor molecule to give the difference spectra.

Figure 7: Spectral changes as expected in the Protoporphyrin IX spectrum

Wavelength

Absorbance

ΔA

ΔA

480nm 600nm


52

1.1 Weak association category: the difference spectra for duroquinone (Figure 11),

quinine (Figure 12) and caffeine (Figure 13) and the Benesi-Hildebrand plots (Figure 14)

for all three compounds are shown below. The difference spectra indicate complex

formation, and Table 3 below shows the dissociation constants (Kd) of the molecules in

the weak association category. The Kd values were obtained from the Benesi-Hildebrand

(B-H) plots.

1.1.1 Duroquinone:

Figure 11: Difference spectra for duroquinone-PPIX complexes, and duroquinone

structure

-0.03

-0.02

-0.01

0

0.01

0.02

500 550 600 650

∆A

nm

Duroquinone-PPIX difference spectra 86.7 mM

71.8 mM

68.7 mM

50.6 mM

47.0 mM

34.4 mM

29.8 mM

23.5 mM


53

1.1.2 Quinine:

Figure 12: Difference spectra for quinine-PPIX complexes and quinine structure

To obtain the correct difference spectra for duroquinone and quinine, further subtraction

of the background absorption in the region 500-650nm was done. It is worth noting that

this further subtraction process may introduce some random errors or greater noise to the

calculated data.

-0.06

-0.04

-0.02

0

0.02

0.04

500 550 600 650ΔA

nm

Quinine-PPIX difference spectra

10mM

20 mM

30 mM

40 mM

60 mM

80 mM


54

1.1.3 Caffeine:

Figure 13: Difference spectra for caffeine-PPIX complexes and caffeine structure

For caffeine, there was no background absorption in the region under study (500-650nm).

Thus, there was no need to subtract any background caffeine spectrum. However, the low

signal to noise ratio obtained for the PPIX-caffeine complexes is entirely due to a very

weak affinity for PPIX-caffeine complexation, since no other significant random errors

were introduced in the subtraction process.

-0.015

-0.01

-0.005

0

0.005

0.01

0.015

0.02

500 550 600 650

ΔA

nm

Caffeine-PPIX difference spectra

100 mM

90 mM

80 mM

70 mM

50 mM

30 mM

20 mM

10 mM


55

Figure 14: Benesi-Hildebrand plots for Duroquinone, Quinine and Caffeine

Table 3: Dissociation constants for molecules in the weak association category

Mean Kd (mM) SD

Quinine 43.3 1.7

Duroquinone 102.4 18.2

Caffeine 953.1 -*

*the Kd value obtained for caffeine indicates very weak complexation, and the Kd value

cannot be determined accurately using the B-H mathematical model, nor by any other

method at hand.

0

1

2

3

4

5

6

7

8

0 0.02 0.04 0.06 0.08 0.1 0.12

[M]/ΔA

[M]

B-H plots for weak association molecules

DQ/ΔA517nm

QN/ΔA512nm

CAF/ΔA516nm


56

Discussion for weak association molecules:

Duroquinone has a simple structure made up of an aromatic 1,4-benzoquinone ring with

2,3,5,6-tetramethyl substitution on the ring (Figure 11). This compound was chosen

because of its potential for π-π hydrophobic interactions. For example, complex

formation between duroquinone and a tetraphenylporphyrin derivative has been

demonstrated using optical spectrophotometry 50

. Caffeine was also chosen because it

has a xanthine ring structure which is a conjugated planar ring structure. Due to its

structure, caffeine is able to form π-π complexes with some DNA intercalators 58

.

Quinine was chosen because it is structurally related to chloroquine. However based on

the results of this experiment, quinine falls under the weak association molecule category.

Although quinine has a quinoline ring which has potential for π-π hydrophobic

interaction, the weak association observed could be as a result of steric hindrance due to

the presence of the tricyclic substituent on the quinoline ring. The steric hindrance may

prevent the correct conformation required for successful complexation with PPIX.

The Ka values were determined from the B-H plots (Figure 14) according to the relation:

Ka =slope/intercept where the latter values are taken from the B-H plot. The values

reported are Kd values, each of which is equal to the inverse of the respective Ka values.

Generally, the lower the Kd value, the stronger is the association. Quinine showed a

stronger association than duroquinone. From the Kd values (Table 3) caffeine showed a

very high Kd with very high associated errors. This implies a very weak association that

is probably beyond the limit of what the B-H model can measure accurately.

Thus, the xanthine ring in caffeine may not be suitable for the hydrophobic interaction

that we anticipate. Duroquinone is mostly neutral with no possible charges on the ring in


57

nonaqueous systems where there is no possibility of forming the duroquinone

semiquinone. Caffeine is also very weakly basic (with a pKa = 1.39) requiring a strong

acid to protonate it 59

. Therefore caffeine is 100% neutral in vivo and this may explain its

limited binding with PPIX.

π-π binding is the force of attraction that occurs between two aromatic ring systems when

their π-electron systems overlap significantly in face-to-face complexes, and very close

proximity is required for this interaction. Such aromatic ring systems may be required for

the complexation reactions with PPIX since duroquinone was able to form weak

complexes while caffeine (with its less aromatic system) formed much weaker

complexes. The weak complexes formed by duroquinone may be due to steric hindrance

from the substituted tetra-methyl groups, which may limit the extent of overlap by the π

system of electrons. In the case of quinine we observe that in addition to the π-π bonding,

a favorable structural conformation may also be required.

Weak versus Strong association: the molecules studied were broadly classified into two

categories based on their observed association constants. Comparing the difference

spectrum of duroquinone (Figure 11) with that of chloroquine (Figure 15), both

molecules produced ∆A values of approximately 0.02 but at different concentration

ranges. For significant complexation, the chloroquine concentration was in the

micromolar (µM) range while the duroquinone concentration was in the millimolar (mM)

range. This shows that the chloroquine acceptor was able to achieve the same effect as

the duroquinone acceptor at a much lower concentration range, thus indicating a stronger

association with PPIX. Other acceptor molecules were classified based on this criterion.


58

The major contrast between the weak and the strong association categories is the

presence of the basic nitrogen in the strong association acceptor molecules. With the

potential proton donor-acceptor interaction in play, there could be electrostatic

interactions between the negatively charged carboxylic acid on PPIX and the positively

charged nitrogen on the acceptor molecules with strong association. It is clear that there

can be proton exchange between PPIX and the basic nitrogen even in the nonaqueous

systems of this study. Thus, we propose that three structural features may be important

for the acceptor molecules: an aromatic ring system, basic nitrogen centres as well as a

favorable conformation for the complexation reaction.

1.2 Strong association category:

1.2.1 Difference Spectra:

The difference spectra shown below were obtained by subtracting the PPIX spectrum

from the respective PPIX-complex spectra. For chloroquine, there was no significant

background absorption in the 500 – 650 nm region, and there was no need for further

subtraction of the underlying absorption. Whereas for other molecules: amodiaquine,

mefloquine, quinacrine and pyronaridine, there was a need to further subtract the

underlying absorption of the respective molecules in order to obtain the correct difference

spectra (Figure 10).

The difference spectra obtained for each molecule showed unique patterns (Figures 15-

19). This demonstrates unique structural features for each of the acceptor molecules in

their respective complexes with PPIX. For instance, the difference spectra for MFQ

(Figure 17) had more maxima and minima at specific wavelengths that could be used in

the Hill plot analysis (see below) compared to all other molecules.


59

Chloroquine, the lead compound in this series of experiments, is currently being used

clinically in the treatment of cutaneous porphyria. Therefore, the complex formation

demonstrated here between chloroquine and PPIX in a hydrophobic model strongly

supports its current clinical use in the treatment of porphyria cutanea tarda (PCT), given

the very hydrophobic nature of the skin’s stratum corneum. It is logical to propose that

some other molecules that are structurally related to chloroquine may also form similar

complexes with PPIX and therefore be useful in the treatment of porphyria.


60

Figure 15: Difference spectra for chloroquine-PPIX complexes and chloroquine

structure

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

500 550 600 650∆A

nm

Chloroquine-PPIX difference spectra

34 μM

137 μM

265 μM

530 μM

1326 μM

5305 μM


61

Figure 16: Difference spectra for amodiaquine-PPIX complexes and amodiaquine

structure

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

500 550 600 650

ΔA

nm

Amodiaquine-PPIX difference spectra

10745 µM

6447 µM

4298 µM

2149 µM

1289 µM

859.6 µM

429.8 µM

214.9 µM

128.9 µM

85.96 µM

42.98 µM


62

Figure 17: Difference spectra for mefloquine-PPIX complexes and mefloquine structure

-0.08

-0.04

0

0.04

0.08

500 550 600 650ΔA

nm

Mefloquine-PPIX difference spectra

3673 µM

734.6 µM

367.3 µM

183.0 µM

122.0 µM

73.46 µM

36.73 µM

24.50 µM

14.70 µM


63

Figure 18: Difference spectra for quinacrine-PPIX complexes and quinacrine structure

-0.04

-0.02

0

0.02

0.04

500 550 600 650

ΔA

nm

Quinacrine-PPIX difference spectra

17.09 µM

34.18 µM

68.36 µM

102.5 µM

170.9 µM

341.8 µM

683.6 µM

1709 µM

3418 µM

5127 µM

8545 µM


64

Figure 19: Difference spectra for pyronaridine-PPIX complexes and pyronaridine

structure

-0.025

-0.015

-0.005

0.005

0.015

0.025

0.035

0.045

500 550 600 650

∆A

nm

Pyronaridine-PPIX difference spectra 916 µM

458 µM

229 µM

152 µM

91.6 µM

45.78 µM

30.53 µM

18.32 µM

9.16 µM

3.66 µM


65

1.2.2 Binding Curves:

The changes in absorbance values (ΔA values) as read from the difference spectra at

specific wavelengths were used to further analyse the nature of the complexes using

appropriate graphical and mathematical models. The same ΔA values for each molecule

were used to plot the binding curves, the Hill plots and the Scatchard plots.

The binding curves below show the trend of the changes in optical absorbance (ΔA

values) as the acceptor molecules’ concentrations were increased for each case. The ΔA

values increased as the acceptor molecule concentration increased and for most of the

curves (except pyronaridine) the ΔA value increased until it reached a plateau. At the

highest concentrations, the ΔA value even dropped below the maximum in several cases.

This is most likely an indication of less 1:1 complex formation occurring at very high

acceptor concentration. Alternatively, other types of complexes may form at higher

acceptor concentrations such as self-association between the acceptor molecules or

perhaps 2:1 binding of acceptor molecules to the PPIX molecule. For pyronaridine

(PYR), the ΔA values steadily increased with increase in PYR concentration without

reaching a maximum. We could not study the behaviour at higher PYR concentration

ranges than those given in Figure 25 below because of the large underlying PYR

absorption that will occur at those higher concentrations. For instance, the underlying

absorption in the region being studied was ~1.5 at wavelength 480nm at the highest

concentration of pyronaridine studied (0.916mM) as shown in Figure 10. This makes it

difficult to subtract the underlying absorption without introducing significant systematic

errors to the calculated difference spectra for the PPIX-pyronaridine case, in particular.

The Beer-Lambert Law is unreliable for absorbance values greater than 1.0. Furthermore,


66

it is increasingly inaccurate for absorbance values greater than 2.0 (with errors of ~5%).

The very large absorbance values are entirely due to the high concentrations of

pyronaridine that absorbs very strongly (Figure 10). The binding curves for the acceptor

molecules that formed strong complexes with PPIX are as shown in Figures 20-25 below.

Figure 20: Binding curves for chloroquine at four different wavelengths

Binding curves: Chloroquine

[CQ]

0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014

-0.04

-0.03

-0.02

-0.01

0.00

0.01

0.02

0.03

499nm

548nm

516nm

534nm


67

Figure 21: Binding curves for amodiaquine at four different wavelengths (n=3)

Binding Curves: Amodiaquine n=3

[AMQ]

0.000 0.002 0.004 0.006 0.008 0.010 0.012

-0.03

-0.02

-0.01

0.00

0.01

0.02

0.03

502nm

521nm

550nm

536nm

Figure 22: Binding curves for amodiaquine at four different wavelengths (n=6)

Binding curves: Amodiaquine n=6

[AMQ]

0.000 0.002 0.004 0.006 0.008 0.010 0.012

A

-0.03

-0.02

-0.01

0.00

0.01

0.02

0.03

521nm

502nm

536nm

550nm


68

Repetition of amodiaquine experiments were done in triplicates (n=3) and in n=6

determinations. Figures 21 and 22 show the binding curves for the two sets of

determinations. We observe that there was no apparent improvement in the quality of the

data in terms of the error bars when the number of measurements was increased from n=3

to n=6. Therefore, triplicate measurements were adopted for the other acceptor molecules

as well. Finally, this phenomenon will be explained in Section 1.2.4 during the

discussion of the Hill plots where the amodiaquine error bars showed a similar behaviour

for n=3 and n=6 experimental data.

Figure 23: Binding curves for mefloquine at four different wavelengths

Binding Curves: Mefloquine

[MFQ]

0.000 0.001 0.002 0.003 0.004

-0.08

-0.06

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

535nm

548nm

516nm

633nm


69

Figure 24: Binding curves for quinacrine at four different wavelengths

Binding Curves: Quinacrine

[QCR]

0.000 0.002 0.004 0.006 0.008 0.010

-0.04

-0.03

-0.02

-0.01

0.00

0.01

0.02

0.03

0.04

501nm

520nm

550nm

535nm

Figure 25: Binding curves for pyronaridine at four different wavelengths

Binding curves: Pyronaridine

[PYR]

0.0000 0.0002 0.0004 0.0006 0.0008 0.0010

Delta A

-0.03

-0.02

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

498nm

519nm

535nm

550nm


70

1.2.3 Scatchard Plots:

The shape and nature of the Scatchard plots can be used to determine interactions

between multiple binding sites in donor-acceptor complexes. When this model is used in

the study of drug-receptor binding, a linear plot shows that there is only one binding site

while non-linear curves imply multiple binding sites 56

. However, in the context of the

complexation reactions in this study, a non-linear curve may imply that there are multiple

conformations for both the PPIX and the acceptor molecules that are favourable for the

complex formation. In other words, there may be different kinds of specific

intermolecular interactions between PPIX and each of the acceptor molecules in this

study. In terms of the potential conformational changes, PPIX is amenable to rotation of

the carboxylic acid groups as propionic acid moieties. However, many of the respective

acceptor molecules such as chloroquine are amenable to more conformational degrees of

freedom; and this is a major factor in the different behaviours observed for the various

acceptor molecules, particularly in the Scatchard plots (Figures 26-29).

The chloroquine data gave non-linear plots that always curved downwards (Figure 26) at

higher chloroquine concentrations. This most likely implies that more than one form of

the chloroquine-PPIX complex is being formed: either through self-association of

chloroquine or through formation of a ternary 2:1 chloroquine: PPIX complex. Higher

order complexes such as 3:1 relative stoichiometries are very unlikely due to steric

restrictions.

Amodiaquine: the Scatchard plot for amodiaquine, similar to chloroquine, is also non-

linear curving downwards (Figure 27). This also implies that there may be more than one

form for the amodiaquine-PPIX complexes being formed.


71

Quinacrine: the Scatchard plot for quinacrine appears to be mostly linear for the data

measured at 535 nm (in Figure 28) over most of the concentration range. This may imply

that there is likely only one form for the quinacrine-PPIX complexes. In other words, it

may be that the quinacrine structure is neither amenable to significant conformational

changes nor self-association. However, at the higher end of the fractional binding (v) on

the abscissa axis, we observed a hook-like shape on the Scatchard plot (see circle on

Figure 28). A similar pattern was also observed in the binding curve (Figure 24) where

the ΔA values drops after reaching a maximum. This may be due to complex interactions

such as self-association of quinacrine molecules or formation of ternary (2:1) complexes

occurring for higher concentrations of the acceptor molecule.

Mefloquine: the Scatchard plot for mefloquine also showed a hook curving upwards at

the higher end of fractional binding (see circle on Figure 29). This may be as a result of

some complex interactions occurring at the high concentration end such as self-

association of mefloquine or formation of the ternary (2:1) complex postulated just above

for quinacrine.


72

Figure 26: Scatchard plots for chloroquine at 516nm in triplicate measurements

Figure 27: Scatchard plots for amodiaquine at 521 nm in triplicate measurements

Scatchard plots: Amodiaquine

v

0.0 0.2 0.4 0.6 0.8 1.0 1.2

v/[AMQ]

0

2000

4000

6000

8000

10000

data set 1 521nm

data set 2 521nm

data set 3 521nm

Scatchard Plots: Chloroquine

v

0.0 0.2 0.4 0.6 0.8 1.0 1.2

v/[CQ]

0

10000

20000

30000

40000

50000

60000

70000

data set 1 516nm

data set 2 516nm

data set 3 516nm


73

Figure 28: Scatchard plots for quinacrine at 535nm in triplicate measurements

Scatchard plots: Quinacrine

v

0.0 0.2 0.4 0.6 0.8 1.0 1.2

v/[QCR]

0

2000

4000

6000

8000

10000

12000

14000

16000

data set 1 535nm

data set 2 535nm

data set 3 535nm

Figure 29: Scatchard plots for mefloquine at 516 nm in triplicate measurements

Scatchard plots: Mefloquine

v

0.0 0.2 0.4 0.6 0.8 1.0 1.2

v/[MFQ]

0

5000

10000

15000

20000

25000

data set 1 516nm

data set 2 516nm

data set 3 516nm


74

1.2.4 Hill plots:

The Hill plots below (Figures 30-33) show the binding of each molecule, and the

dissociation constants can easily be determined by reading of the intercept of the plot on

the x-axis= Log Kd based on the Hill equation.

Figure 30: Hill plot for chloroquine (516nm)

Hill plot -Chloroquine

Log [CQ]

-6.0 -5.5 -5.0 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0

Log[v/(1-v)]

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Chloroquine (516nm)


75

Figure 31: Hill plots for amodiaquine (521nm) (n=3 determination)

Figure 32: Hill plot for amodiaquine (521nm) (n=6 determination)

Repetition of the amodiaquine experiments was done for the n=6 determination in order

to determine the effect of the number of repeated measurements on the signal to noise

ratio and the magnitude of the error bars in the Hill Plot, presented just above. This

comparison was also made for the measurement of the binding curves for PPIX–

Hill plot -Amodiaquine (n=3 determination)

Log[AMQ]

-4.5 -4.0 -3.5 -3.0 -2.5 -2.0

Log[v/(1-v)]

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

AMQ(n=3) 521nm

Hill plot -Amodiaquine (n=6 determination)

Log[AMQ]

-4.5 -4.0 -3.5 -3.0 -2.5 -2.0

Log[v/(1-v)]

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

AMQ (n=6) 521nm


76

amodiaquine presented earlier in Figures 21 and 22. From the two Hill plots for n=3 and

n=6 (Figures 31 and 32) shown above, we find that there was no apparent improvement

in the magnitude of the errors in the AMQ n=6 determination. This means that the

random errors observed in the determinations are inherent in the series of subtraction

processes involved in obtaining the difference spectra. Apart from subtraction of the

PPIX reference spectrum, it was necessary to further subtract the tailing underlying

absorption of the acceptor molecules (with the exception of chloroquine). The double

subtraction process was most likely responsible for the random errors that were observed.

Thus, for these particular experiments with significant background absorption by most of

the acceptor molecules at high concentrations, triplicate measurements were chosen for

the experimental determinations for all of the acceptor molecules being tested.

Figure 33: Hill plots for mefloquine (516nm) and quinacrine (535nm)

Hill Plots: Mefloquine / Quinacrine

Log [M]

-5.0 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5

Log [v/(1-v)]

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Mefloquine (516nm)

Quinacrine (535nm)


77

Table 4: Dissociation constants for molecules in the strong association category

Mean Kd (µM) SD

Mefloquine 39.2 6.1

Chloroquine 100.4 39.4

Quinacrine 108.6 23.2

Amodiaquine 378.8 30.2

Pyronaridine -* -*

*No quantitative data was obtained for pyronaridine due to the strong background

absorption.

Dissociation constants (Kd) values: the Kd values for the strong association molecules are

shown above in Table 4. The values are in the µM range. Mefloquine had the lowest Kd

value implying the strongest association of all the four molecules. Mefloquine is also the

only compound that showed a stronger association than chloroquine. Quinacrine had a

slightly higher Kd values, meaning slightly weaker association than chloroquine.

Meanwhile, amodiaquine showed the weakest association. Overall, the structural features

required for the complexation in this study seem to be specific for chloroquine and

structurally related molecules.


78

1.3 Molecular interactions and important structural features:

In line with one of the objectives of this research, it is important to identify the key

structural features that are important for the complex formation. In addition to the

hydrophobic interactions between the aromatic ring system on PPIX and the aromatic

ring systems on the acceptor molecules, proximity between the basic nitrogen (N) on the

acceptor molecules and the oxygen (O) of the carboxylic acid on the PPIX structure can

lead to electrostatic interactions between positive and negative charges when proton

transfer occurs. This electrostatic attraction between the positively charged N and

negatively charged carboxylic acid O may be responsible for the relatively strong

association observed for the chloroquine, amodiaquine, mefloquine and quinacrine

acceptor molecules.

Table 5 below shows the pKa values due to the potential protonation of the most basic N-

atom(s) on each of the acceptor molecules. Note that only pyronaridine has two very

basic N-atoms and hence it has the potential to be doubly ionized.

Table 5: pKa values of the acceptor molecules (most basic nitrogens)

Acceptor molecule pKa1 pKa2

Amodiaquine60

- 8.1

Quinacrine61

- 10.2

Chloroquine60

- 10.1

Quinine53

- 8.5

Mefloquine53

- 8.6

Pyronaridine62

9.88 10.3


79

The figures (Figures 4, 34-37) below show the potential structural interactions between

PPIX and the acceptor molecules with both hydrophobic and electrostatic attractions in

play as discussed above.


a 1:1 complex


80

Figure 34: Theoretical superimposition of protoporphyrin IX and quinacrine structures in

a 1:1 complex

Figure 35: Theoretical superimposition of protoporphyrin IX and amodiaquine structures

in a 1:1 complex


81

Figure 36: Theoretical superimposition of protoporphyrin IX and mefloquine structures

in a 1:1 complex

In the case of pyronaridine, the results showed a very strong indication for binding as

shown on the binding curves and difference spectra (Figures 19 and 25). Here, we

observed increasing ΔA values that did not reach a plateau value. Even though we could

not deduce the dissociation constant value for pyronaridine, the results showed a very

strong indication for binding with PPIX. The strong binding observed might be as a result

of unique structural features of pyronaridine. Pyronaridine is a benzonaphthyridine

derivative (Figure 19) with a total of four basic nitrogen atoms in its structure of which

the two nitrogen atoms in the pyrrolidine rings are the most basic. Upon superimposition

on the PPIX ring structure (Figure 37), these two most basic N atoms on the pyronaridine

structure can engage simultaneously with the two negatively charged O atoms on the two

carboxylic acids on the PPIX via induced electrostatic attraction.


82

Figure 37: Theoretical superimposition of protoporphyrin IX and pyronaridine structures

in a 1:1 complex

We propose that PYR with its two very basic nitrogen atoms showed such a strong

binding affinity indication because of the added electrostatic attractions that can

effectively engage both carboxylic acid groups on PPIX in addition to the hydrophobic

interactions between the aromatic ring systems. Thus, the electrostatic attraction

mechanism appears to be very important for favourable complexation with PPIX.


83

1.4 Unique PPIX complexes with other acceptor molecules: amitriptyline and

chlorpromazine:

In order to explore further the structural features required for the PPIX complexation, we

screened other compounds that are outside of the chloroquine structural domain. These

molecules, amitriptyline and chlorpromazine, were selected because of the presence of

ring systems in their structures and the presence of one very basic nitrogen in their side

chains. In the case of amitriptyline, the ring system (dibenzocycloheptane) is clearly not

perfectly aromatic. However, we hypothesized that the chlorpromazine ring may be

sufficiently aromatic for the facilitation of some complexation.

The measured difference spectra for the two molecules are shown in Figures 39 and 40. It

is interesting to note that the nature of the results obtained was totally different, in fact

opposite to the expected results. Instead of peak broadening which we observed for the

chloroquine related compounds, we saw a narrowing of the PPIX spectrum which

resulted in a different shape of the difference spectra. The difference spectra showed that

there is a unique interaction or complex formation between PPIX and each of

amitriptyline and chloropromazine, but definitely not the type of complexes that we are

interested in. The band narrowing observed may be as a result of an electrochromic shift

due to an electric field effect between the negatively charged PPIX and the positively

charged basic acceptor molecules as these unique complexes are formed. In other words,

the interaction involved purely electric field effects with no significant ∏-∏ bonding. A

proposed example of such an intermolecular interaction is represented in Figure 38 just

below. In this electrochromic interaction, there is no significant overlapping of the ∏-

electronic ring systems. The electrochromic shift is defined as a form of electric field


84

effect which may cause changes (such as band narrowing) in the spectrum of PPIX in the

presence of electric charges from the complexing molecule. The evidence for band-

narrowing is clearly present in the difference spectra (Figures 39 and 40), shown below.

Figure 38: Superimposition of PPIX and chlorpromazine in a possible electrochromic

interaction complex


85

Figure 39: Difference spectra for amitriptyline-PPIX complexes and amitriptyline

structure

-0.05

-0.04

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

500 550 600 650∆A

nm

Amitriptyline-PPIX difference spectra

100 mM

80 mM

60 mM

40 mM

20 mM

10 mM

5.0 mM


86

Figure 40: Difference spectra for chlorpromazine-PPIX complexes and chlorpromazine

structure

-0.06

-0.05

-0.04

-0.03

-0.02

-0.01

0

0.01

0.02

500 550 600 650∆A

nm

Chlorpromazine-PPIX difference spectra

150 mM

90 mM

75 mM

60 mM

30 mM

15 mM

7.5 mM


87

2.0 Triton X-100® model:

Figure 9: Triton X-100® structures

In the Triton X-100® model (Figure 9), the focus is to target the hydrophobic core of the

micelles and determine whether or not complexation will occur in this hydrophobic

environment. For the two molecules tested, amodiaquine and chloroquine, the difference

spectra are as shown in Figures 41 and 42 respectively. The behaviour of the difference

spectra is similar to the behaviour observed in the homogenous solvent system described

in detail previously. This implies that the form of complexes we are interested in can also

be formed in such a hydrophobic environment. The only limitation of the micellar model

is that we could not quantify the concentration of chloroquine and amodiaquine that both

successfully partition into the hydrophobic core. This partitioning was proven by the

observation of significant optical absorption difference spectra for these molecules. We

can only report the amount of chloroquine or amodiaquine dissolved in the aqueous

phase. Also because of the complex structure of the micellar partitioning system, higher

concentrations of the molecules (mM) were required compared to the concentrations used

in the homogenous solvent system of acetone/dichloromethane. It is also important to

note that for the micellar systems, it was necessary to use the salt form to dissolve each of


88

the acceptor molecules in the aqueous phase first at sufficiently high concentration to

drive the partitioning into the hydrophobic micellar interior through deprotonation at the

aqueous interface.

Figure 41: Difference spectra for amodiaquine-PPIX in Triton X-100®

Figure 42: Difference spectra for chloroquine-PPIX in triton X-100®

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

0.04

0.05

500 520 540 560 580 600 620 640

Amodiaquine-PPIX difference spectra in Triton X-100®

20mM

15mM

10mM

5.0mM

2.0mM

1.0mM

0.5mM

-0.02

-0.015

-0.01

-0.005

0

0.005

0.01

0.015

0.02

500 550 600 650

Chloroquine-PPIX difference spectra in Triton X-100®

0.036mM

1.0 mM

2.0 mM

4.0 mM

10 mM

20 mM


89

3.0 Bovine serum albumin:

PPIX spectrum in the presence of BSA: lower absorbance

From the spectrum of PPIX reference in the presence of BSA (Figure 43) we observe that

the absorbance values were low compared to the values obtained at the same

concentration in homogenous solvents (Figure 44). This was as a result of poor

incorporation of PPIX at the aqueous interface of the BSA protein in the hydrated BSA

protein system. The main difficulty is that PPIX is insoluble in the aqueous phase itself.

It should also be noted that the incorporation of the PPIX in BSA was a slow process that

took a number of days (three days), and yet complete incorporation was not achieved.

There was also a slight shift to the red in the PPIX spectrum peaks in the BSA model

compared to the acetone-dichloromethane homogenous solvent system. For instance the

first Q band peak was shifted from 502 nm in the homogenous systems to 510 nm in the

BSA model. Furthermore, there was a broadening of the peaks of the PPIX spectrum in

the BSA model as shown in Figure 43. This may be a result of some complex formation

between the PPIX and BSA, (since this happens before the addition of CQ). The

broadening of the spectrum will also contribute to the reduced peak amplitudes of the

spectrum.


90

Figure 43: PPIX spectrum (36µM) in the presence of BSA

Figure 44: PPIX Spectrum (36µM) in acetone/dichloromethane solvent system

*Random changes occurs at about 670nm in the homogeneous solvent systems but does

not appear in the presence of BSA.

0

0.1

0.2

0.3

0.4

0.5

480 530 580 630 680

Ab

sorb

ance

nm

PPIX spectrum (36µM) in the presence of BSA

0

0.1

0.2

0.3

0.4

0.5

480 530 580 630 680

Ab

sorb

ance

nm

PPIX Spectrum (36µM) in Acetone/Dichloromethane Solvent system

*


91

Changes in the PPIX spectrum in the presence of BSA and Chloroquine

The difference spectrum of the complexation of PPIX with chloroquine in the presence of

BSA did not show any well characterized spectrum in the region 500-650 nm. This

implies that there was no evident complex formation in this model. If there were any

PPIX-chloroquine complex formation, some changes in the PPIX spectrum should have

occurred.

However, one very important piece of information that we could draw out of the BSA

experiments is the fact that we did not observe the “random” changes in the PPIX

spectrum around 670 nm region that were observed in previous experiments carried out

in the homogenous solvent model. In other words, the small peak at 670 nm in the

homogenous mixture PPIX spectrum (Figure 44) disappears and flattens out in the

presence of BSA (Figure 43). The absence of these changes may be partly due to the

immobilization of the PPIX structure on the BSA, such that the conformational changes

or rotations that would normally give rise to the “random” changes do not occur. It is the

two pendant propionic acid moieties of PPIX that are subject to the most important

conformational changes in PPIX.

PPIX has been shown to bind with plasma proteins such as albumin 26

. Thus, it could be

reasoned that the binding affinity between PPIX and BSA may be stronger than the

potential affinity between PPIX and chloroquine. Therefore, it may be difficult for

chloroquine to displace PPIX from its binding sites on BSA. Furthermore, this may

explain why there was no evidence for PPIX complexation with chloroquine in the

presence of BSA in an aqueous environment at pH 7.40 in our experiments.


92

Chloroquine in BSA spectrum: chloroquine ionized:

In order to deduce the location of chloroquine in the PPIX-BSA mixture, and to know if

there will be a potential interaction between chloroquine and BSA, changes in the

spectrum of chloroquine were studied at two pH values in the presence of BSA. At pH

9.00 (Figure 45), chloroquine appears in the spectrum as carrying only one charge (with

the ~90% protonation of the more basic N-atom with pKa = 10.1). However at pH 7.40

(also in Figure 45) we observed two peaks indicating two ionized charges on the

chloroquine molecule, which of course would be similar to the spectrum of chloroquine

at pH 7.40 in the pure aqueous system (with no BSA).

Figure 45: Spectrum of chloroquine in the presence of BSA at pH 7.40 and pH 9.00

This implies that at pH = 7.40, that is most relevant to our study, the highly charged

chloroquine molecules remain mostly in the aqueous medium, and this may be one reason

for its limited interaction with the BSA or particularly for the PPIX-BSA mixture. It was

clear that no significant optical absorption difference spectra due to PPIX-chloroquine

complexes were observed in the aqueous BSA system. A second potential explanation for

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

310 360 410 460 510 560

Chloroquine spectrum in the presence of BSA

pH 9.00

pH 7.40


93

the lack of a PPIX-chloroquine difference spectrum in this aqueous BSA system was that

the BSA-immobilized PPIX (binding at hydrophobic sulfhydryl sites) was subject both to

steric factors and to π-electronic factors that made it much less available for facile

complexation with the chloroquine molecules.

Conclusions and Recommendations

94

CHAPTER V: CONCLUSIONS AND RECOMMENDATIONS

Limitation of in-vitro models:

Most binding studies are performed in vitro with diluted concentrations of ligands and

macromolecules. However, these conditions do not completely illustrate conditions that

are found in a cell in vivo. There are multi-component structures such as organelles,

macromolecular complexes and cytoskeletal components which make up the structure of

the cells. Therefore, binding studies of endogenous macromolecules such as PPIX in a

homogenous solvent system model may be an oversimplified model system. Nonetheless,

according to the rationale and hypothesis presented for this study, the hydrophobic

homogenous solvent system is still a very useful model for studying the complexation

reactions in vitro since it will be virtually impossible to measure dissociation constants of

the complexes in the more complicated in vivo conditions.

1.0 Conclusions:

Complex formation has been established between PPIX and a range of acceptor

molecules in the hydrophobic acetone-dichloromethane solvent model. Furthermore, the

respective dissociation constants for the PPIX-complexes have been determined using

appropriate mathematical and graphical models.

Higher dissociation constants (mM range), implying weaker complexes, were obtained

for simpler molecules such as duroquinone, quinine and caffeine. While much lower

dissociation constants (µM range), implying stronger association, were obtained for PPIX

complexes of chloroquine, mefloquine, quinacrine, and amodiaquine. Pyronaridine

showed a very strong indication for PPIX-binding. However, the exact dissociation

constant could not be determined due to its very large background absorption.


95

In terms of the structural features required for the PPIX complex formation, an aromatic

ring system is necessary for the complexation, but the hydrophobic bonding alone is not

sufficient for stronger binding. In the case of duroquinone, weak complexation may be as

the result of tetra-methyl substitution on the quinone ring, thus limiting proper

overlapping of the π systems of electrons with those of PPIX. As it was observed for

chloroquine and other molecules within the strong association category, basic nitrogen

centres are very important for stronger binding due to the added electrostatic interactions

that can then occur in the PPIX complexes. This importance of the basic nitrogen centre

is demonstrated further by the strong binding indication observed for pyronaridine which

has two very basic basic nitrogen centres.

In the Triton X-100® model, complex formation with PPIX was established for

chloroquine and amodiaquine in the hydrophobic core of the micelles. Our results show

that PPIX forms complexes in this particular hydrophobic model. Such a hydrophobic

model is somewhat similar to the highly hydrophobic nature of the skin for instance.

However, due to inability to quantify the exact amount of chloroquine or amodiaquine

that partitions into the micellar core, dissociation constants could not be determined in the

micellar system.

Finally, there was no observable complex formation between PPIX and chloroquine in

the hydrated bovine serum albumin (BSA) model system. This might be the result of the

competing stronger interaction between PPIX and the BSA such that chloroquine cannot

displace PPIX from its hydrophobic binding sites on BSA.


96

2.0 Recommendations:

Chloroquine forms relatively strong complexes with PPIX, and this is believed to be its

mechanism of action in porphyria therapy involving enhanced elimination or excretion of

excess PPIX in humans. Looking forward, there may be some promise for other

molecules examined in this study that are structurally related to chloroquine that can also

form strong complexes with PPIX in porphyria therapies. Among the compounds tested,

mefloquine and pyronaridine are particularly promising.

Since all the drug molecules in this study are already being used clinically, although for

different indications, animal toxicity preclinical studies will not normally be required for

these already existing drugs in porphyria therapy.

References

97

References

1. Desnick RJ. The porphyrias. In: Dennis L. Kasper, Tinsley Randolph Harrison, editors.

Harrison's principles of internal medicine. 16th edition ed. New York: McGraw-Hill,

Medical Pub. Division; 2005.

2. Puy H, Deybach JC. Heme biosynthesis and excretion of porphyrins. In: J. Rodés, J. -P

Benhamou, A. Blei, J. Reichen, M. Rizzetto, editors. Textbook of hepatology. third

edition ed. Wiley-Blackwell; 2008.

3. Poh-Fitzpatrick MB. Porphyrias: Photosensitivity and phototherapy. Methods Enzymol

2000;319:485-93.

4. Pietrangelo A. The porphyrias: Pathophysiology. Intern Emerg Med 2010 Oct;5 Suppl

1:S65-71.

5. Thunell S, Pomp E, Brun A. Guide to drug porphyrogenicity prediction and drug

prescription in the acute porphyrias. Br J Clin Pharmacol 2007 Nov;64(5):668-79.

6. Herrero C, Vicente A, Bruguera M, Ercilla MG, Barrera JM, Vidal J, Teres J, Mascaro

JM, Rodes J. Is hepatitis C virus infection a trigger of porphyria cutanea tarda?

Lancet 1993 Mar 27;341(8848):788-9.

7. Andersson C, Innala E, Backstrom T. Acute intermittent porphyria in women: Clinical

expression, use and experience of exogenous sex hormones. A population-based

study in northern sweden. J Intern Med 2003 Aug;254(2):176-83.

8. Genetics Home Reference: Porphyria [Internet]US: US National Library of Medicine;

c2009 [cited 2011 03/26]. Available from:

http://ghr.nlm.nih.gov/condition=porphyria.

9. Poblete-Gutierrez P, Wiederholt T, Merk HF, Frank J. The porphyrias: Clinical

presentation, diagnosis and treatment. Eur J Dermatol 2006 May-Jun;16(3):230-40.

10. First line tests -American Porphyria Foundation [Internet]United States: American

Porphria Foundation; c2010 [cited 2012 May 14]. Available from:

http://www.porphyriafoundation.com/testing-and-treatment/testing-for-

porphyria/first-line-tests.

11. Cappellini MD, Brancaleoni V, Graziadei G, Tavazzi D, Di Pierro E. Porphyrias at a

glance: Diagnosis and treatment. Intern Emerg Med 2010 Oct;5 Suppl 1:S73-80.

12. Clinuvel Pharmaceuticals - Clinical trial results [Internet]Melbourne Australia:

Clinuvel Pharmaceuticals Ltd; c2012 [cited 2012 May 14]. Available from:

http://www.clinuvel.com/en/scenesse/clinical-results.

References

98

13. Clinuvel Pharmaceuticals -ongoing epp program [Internet]Melbourne, Australia:

Clinuvel Pharmaceuticals; c2012 [cited 2012 May 14]. Available from:

http://www.clinuvel.com/en/scenesse/ongoing-epp-program.

14. Song X, Mosby N, Yang J, Xu A, Abdel-Malek Z, Kadekaro AL. Alpha-MSH

activates immediate defense responses to UV-induced oxidative stress in human

melanocytes. Pigment Cell Melanoma Res 2009 Dec;22(6):809-18.

15. Harms J, Lautenschlager S, Minder CE, Minder EI. An alpha-melanocyte-stimulating

hormone analogue in erythropoietic protoporphyria. N Engl J Med 2009 Jan

15;360(3):306-7.

16. Clinuvel, FDA agrees in principle to phase III study with Scenesse®

(Aflamelatonide) [Internet]Melbourne, Australia: Clinuvel Pharmaceuticals; c2012

[cited 2012 May 14]. Available from: http://www.clinuvel.com/en/investors/news-

publications/announcements/2012/clinuvel-fda-agree-in-principle-to-phase-iii-study-

with-scenesse%C2%AE-afamelanotide.

17. Hydroxychloroquine and Phlebotomy for treating of porphyria cutanea tarda

[Internet]United States: US National Institute of health; c2012 [cited 2012 May 14].

Available from:

http://clinicaltrials.gov/ct2/show/NCT01573754?term=porphyria&rank=5.

18. Safety and efficacy of oral deferasirox in patients with porphyria cutanea tarda

[Internet]United States: US National Institute of Health; c2012 [cited 2012 May 14].

Available from:

http://clinicaltrials.gov/ct2/show/NCT01284946?term=porphyria&rank=15.

19. Anderson KE, Bloomer JR, Bonkovsky HL, Kushner JP, Pierach CA, Pimstone NR,

Desnick RJ. Recommendations for the diagnosis and treatment of the acute

porphyrias. Ann Intern Med 2005 Mar 15;142(6):439-50.

20. Mustajoki P, Nordmann Y. Early administration of heme arginate for acute porphyric

attacks. Arch Intern Med 1993 Sep 13;153(17):2004-8.

21. Mustajoki P, Tenhunen R, Tokola O, Gothoni G. Haem arginate in the treatment of

acute hepatic porphyrias. Br Med J (Clin Res Ed) 1986 Aug 30;293(6546):538-9.

22. Mathews-Roth MM. Carotenoids in erythropoietic protoporphyria and other

photosensitivity diseases. Ann N Y Acad Sci 1993 Dec 31;691:127-38.

23. Kostler E, Wollina U. Therapy of porphyria cutanea tarda. Expert Opin

Pharmacother 2005 Mar;6(3):377-83.

24. Ashton RE, Hawk JL, Magnus IA. Low-dose oral chloroquine in the treatment of

porphyria cutanea tarda. Br J Dermatol 1984 Nov;111(5):609-13.

References

99

25. Rotenberg M, Margalit R. Deuteroporphyrin-albumin binding equilibrium. the effects

of porphyrin self-aggregation studied for the human and the bovine proteins.

Biochem J 1985 Jul 1;229(1):197-203.

26. Lamola AA, Asher I, Muller-Eberhard U, Poh-Fitzpatrick M. Fluorimetric study of

the binding of protoporphyrin to haemopexin and albumin. Biochem J 1981 Jun

15;196(3):693-8.

27.Lawrence LW, Broussard LA. Porphyrins and hemoglobin. In: M. L. Bishop, E. P.

Fody, E. P. Scoeff, editors. Clinical chemistry: Principles, procedures and

corrrelations. Fifth Edition ed. Philadelphia: Lippincots Williams and Wilkins; 2004.

28. Sandberg S, Romslo I. Porphyrin-induced photodamage at the cellular and the

subcellular level as related to the solubility of the porphyrin. Clin Chim Acta 1981

Jan 22;109(2):193-201.

29. Nordmann Y, Puy H. Human hereditary hepatic porphyrias. Clin Chim Acta 2002

Nov;325(1-2):17-37.

30. CONNER SK. Systemic lupus erythematosus; a report on twelve cases treated with

quinacrine (atabrine) and chloroquine (aralen). Ann Rheum Dis 1957 Mar;16(1):76-

81.

31. LONDON ID. Porphyria cutanea tarda; report of case successfully treated with

chloroquine. AMA Arch Derm 1957 Jun;75(6):801-3.

32. Rossmann-Ringdahl I, Olsson R. Porphyria cutanea tarda: Effects and risk factors for

hepatotoxicity from high-dose chloroquine treatment. Acta Derm Venereol

2007;87(5):401-5.

33. Saltzer EI, Redeker AG, Wilson JW. Porphyria cutanea tarda. remission following

chloroquine administration without adverse effects. Arch Dermatol 1968

Nov;98(5):496-8.

34. Scholnick PL, Epstein J, Marver HS. The molecular basis of the action of chloroquine

in porphyria cutanea tarda. J Invest Dermatol 1973 Oct;61(4):226-32.

35. Sullivan DJ,Jr, Gluzman IY, Russell DG, Goldberg DE. On the molecular mechanism

of chloroquine's antimalarial action. Proc Natl Acad Sci U S A 1996 Oct

15;93(21):11865-70.

36. Constantinidis I, Satterlee JD. UV-visible and carbon NMR studies of chloroquine

binding to urohemin I chloride and uroporphyrin I in aqueous solutions. J Am Chem

Soc 1988;110(13):4391-5.

References

100

37. Goerz G, Bolsen K, Merk H. Influence of chloroquine on the porphyrin metabolism.

Arch Dermatol Res 1985;277(2):114-7.

38. de Verneuil H, Ged C, Boulechfar S, Moreau-Gaudry F. Porphyrias: Animal models

and prospects for cellular and gene therapy. J Bioenerg Biomembr 1995

Apr;27(2):239-48.

39. Amornrat Jearnpipatkul BP. Molecular complexes of quinoline antimalarials with

iron-porphyrin components of protease digested methemoglobin. Chemico

Biological Interactions 1980;33:83-90.

40. Shanley BC, Clarke K, Winzor DJ. Evaluation of the stoichiometry and strength of

chloroquine-porphyrin interactions by difference spectroscopy. Biochem Pharmacol

1985;34(1):141-2.

41. Constantinidis I, Satterlee JD. UV-visible and carbon NMR studies of quinine

binding to urohemin I chloride and uroporphyrin I in aqueous solution. J Am Chem

Soc 1988;110(3):927-32.

42. Brown SB, Shillcock M, Jones P. Equilibrium and kinetic studies of the aggregation

of porphyrins in aqueous solution. Biochem J 1976;153(2):279-85.

43. Margalit R, Shaklai N, Cohen S. Fluorimetric studies on the dimerization equilibrium

of protoporphyrin IX and its haemato derivative. Biochem J 1983 Feb 1;209(2):547-

52.

44. Marchettini N, Valensin G, Gaggelli E. 1H-NMR studies on the self-association of

chloroquine in aqueous solution. Biophys Chem 1990 5;36(1):65-70.

45. Brault D. Physical chemistry of porphyrins and their interactions with membranes:

The importance of pH. J Photochem Photobiol B 1990 Jun;6(1-2):79-86.

46. Brault D, Vever-Bizet C, Kuzelova K. Interactions of dicarboxylic porphyrins with

membranes in relation to their ionization state. J Photochem Photobiol B 1993

Oct;20(2-3):191-5.

47. Cohen S, Margalit R. Binding of porphyrin to human serum albumin. structure-

activity relationships. Biochem J 1990;270(2):325-30.

48. Robertson CA, Evans DH, Abrahamse H. Photodynamic therapy (PDT): A short

review on cellular mechanisms and cancer research applications for PDT. J

Photochem Photobiol B 2009 Jul 17;96(1):1-8.

49. Foster R. Organic charge-transfer complexes. London: Academic press; 1969.

References

101

50. Gonzalez MC. Ground state complexes between 5-(4-carboxyphenyl)-10,15,20-

tritolylporphyrin and benzoquinones. J Prakt Chem /Chem -Ztg 1993;335(6):515-20.

51. Sweetman S, editor. Martindale: The complete drug reference. Electronic version ed.

London: Pharmaceutical press; 2012.

52. Kurth F, Belard S, Basra A, Ramharter M. Pyronaridine: A new 'old' drug on the

verge of entering the antimalarial armamentarium. Expert Rev Anti Infect Ther 2011

Apr;9(4):393-6.

53. Remington: The science and practice of pharmacy. 21st edition ed. Philadelphia:

Lippincott Williams and Wilkins; 2006.

54. McDonnell PA, Opella SJ. Effect of detergent concentration on multidimensional

solution NMR spectra of membrane proteins in micelles. J Magn Reson , Ser B

1993;102(1):120-5.

55. Carter DC, Ho JZ. Structure of serum albumin. In: V. N. Schumaker, editor.

Advances in protein chemistry, volume 45. California: Academic press Inc; 1994.

56. Marshall AG. Biophysical chemistry: Principles, techniques and applications. New

York: John Wiley & Sons, Inc.; 1978.

57. Strauch S, Jantratid E, Dressman JB, Junginger HE, Kopp S, Midha KK, Shah VP,

Stavchansky S, Barends DM. Biowaiver monographs for immediate release solid

oral dosage forms: Mefloquine hydrochloride. J Pharm Sci 2011 Jan;100(1):11-21.

58. Larsen RW, Jasuja R, Hetzler RK, Muraoka PT, Andrada VG, Jameson DM.

Spectroscopic and molecular modeling studies of caffeine complexes with DNA

intercalators. Biophys J 1996;70(1):443-52.

59. Beale JM, Block JH, editors. Wilson and gisvold's textbook of organic medicinal and

pharmaceutical chemistry. 12th Edition ed. Philadelphia: Lippincott Williams &

Wilkins; 2011.

60. Rosenthal P.J., editor. Antimalarial chemotherapy: Mechanisms of action, resistance

and new directions in drug discovery. 1st edition ed. New Jersey: Humana press Inc;

2001.

61. Drug bank Quinacrine DB01103 [Internet] 3.0. Alberta: Genome Alberta; c2012

[cited 2012 June 20]. Available from: http://www.drugbank.ca/drugs/DB01103.

62. Adegoke OA, Babalola CP, Oshitade OS, Famuyiwa AA. Determination of the

physicochemical properties of pyronaridine - a new antimalarial drug. Pak J Pharm

Sci 2006 Jan;19(1):1-6.

Date post:	08-Dec-2021
Category:	Documents
Upload:	others
View:	9 times
Download:	0 times

PORPHYRIN COMPLEXATION: AN APPROACH IN PORPHYRIA …

Documents